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 775 records · Page 43

Supersonic Free-Flight Dynamics Testing in the Stratosphere

A new technique for obtaining stratospheric free-flight dynamics data for atmospheric entry capsules is described. The Stratospheric Projectile Experiment of Entry Dynamics (SPEED) represents a new approach to characterizing the free-flight dynamics of vehicles in the supersonic and transonic regime of flight. The SPEED test architecture leverages a stratospheric balloon and 3D-printed flight system in a novel two-stage configuration to deliver test articles to supersonic conditions in the atmosphere which achieve dynamic similitude with a full-scale vehicle. Designed to address the limitations of existing test facilities, SPEED captures the complete time evolution of a vehicle’s dynamic state while producing a statistically significant number of observed flight trajectories to address the stochastic nature of the wake-driven dynamic stability phenomenon. SPEED was developed over two years at NASA Ames Research Center. The inaugural flight of the SPEED test platform was conducted in the summer of 2024 where scaled capsules of the Mars Sample Return Earth Entry System and Dragonfly entry vehicle were tested. This paper describes the motivation for a new test architecture, operational constraints and achievable flight test envelopes, the design and development of the SPEED concept, and results from the demonstration flight. Techniques for estimating the dynamic aerodynamic characteristics and atmospheric conditions for each test article individually and as a collective are also discussed.

Entry Vehicle↗

A Scalable Gradient Free Method for Bayesian Experimental Design with Implicit Models

Bayesian experimental design (BED) is to answer the question that how to choose designs that maximize the information gathering. For implicit models, where the likelihood is intractable but sampling is possible, conventional BED methods have difficulties in efficiently estimating the posterior distribution and maximizing the mutual information (MI) between data and parameters. Recent work proposed the use of gradient ascent to maximize a lower bound on MI to deal with these issues. However, the approach requires a sampling path to compute the pathwise gradient of the MI lower bound with respect to the design variables, and such a pathwise gradient is usually inaccessible for implicit models. In this paper, we propose a novel approach that leverages recent advances in stochastic approximate gradient ascent incorporated with a smoothed variational MI estimator for efficient and robust BED. Without the necessity of pathwise gradients, our approach allows the design process to be achieved through a unified procedure with an approximate gradient for implicit models. Several experiments show that our approach outperforms baseline methods, and significantly improves the scalability of BED in high-dimensional problems.

Zhang, Jiaxin↗

Contextual Active Online Model Selection with Expert Advice

How can we collect the most useful labels to learn a model selection policy, when presented with arbitrary heterogeneous data streams? In this paper, we formulate this task as a contextual active model selection problem, where at each round the learner receives an unlabeled data point along with a context. The goal is to output the best model for any given context without obtaining an excessive amount of labels. In particular, we focus on the task of selecting pre-trained classifiers, and propose a contextual active model selection algorithm (CAMS), which relies on a novel uncertainty sampling query criterion defined on a given policy class for adaptive model selection. In comparison to prior art, our algorithm does not assume a globally optimal model. We provide rigorous theoretical analysis for the regret and query complexity under both adversarial and stochastic settings. Our experiments on several benchmark classification datasets demonstrate the algorithm’s effectiveness in terms of both regret and query complexity. Notably, to achieve the same accuracy, CAMS incurs less than 10% of the label cost when compared to the best online model selection baselines on CIFAR10.

Liu, Xuefeng↗

Grain growth competition during thin-sample directional solidification of dendritic microstructures: A phase-field study

Here we present the results of a comprehensive phase-field study of columnar grain growth competition in bi-crystalline samples in two dimensions (2D) and in three dimensions (3D) for small sample thicknesses allowing a single row of dendrites to form. We focus on the selection of grain boundary (GB) orientation during directional solidification in the steady-state dendritic regime, and study its dependence upon the orientation of two competing grains. In 2D, we map the entire orientation range for both grains, performing several simulations for each configuration to account for the stochasticity of GB orientation selection and to assess the average GB behavior. We find that GB orientation selection depends strongly on whether the primary dendrite growth directions have lateral components (i.e. components perpendicular to the axis of the temperature gradient) that point in the same or opposite directions in the two grains. Here, we identify a range of grain orientations in which grain selection follows the classical description of Walton and Chalmers. We also identify conditions that favor unusual overgrowth of favorably-oriented dendrites at a converging GB. We propose a simple analytical description that reproduces the average GB orientation selection from 2D simulations within statistical fluctuations of a few degrees. In 3D, we find a similar GB orientation selection as in 2D when secondary branches grow in planes parallel and perpendicular to the sample walls. Remarkably, quasi-2D behavior is also observed even when those perpendicular sidebranching planes are rotated by a finite azimuthal angle about the primary dendrite growth axis as long as the absolute values of those azimuthal angles are equal in both grains. In contrast, when the absolute values of those azimuthal angles differ markedly, we find that unusual overgrowth events at a converging GB are promoted by a high azimuthal angle in the least-favorably-oriented grain. We also find that diverging GBs can be strongly affected by those azimuthal angles, while converging GBs exhibit a weak dependence on those angles. Finally, for diverging GBs, GB orientation is also strongly affected by the relative signs of the lateral components of the primary dendrite growth directions in both grains.

36 MATERIALS SCIENCE↗

Particle filtering with path sampling and an application to a bimodal ocean current model

This paper introduces a recursive particle filtering algorithm designed to filter high dimensional systems with complicated non-linear and non-Gaussian effects. The method incorporates a parallel marginalization (PMMC) step in conjunction with the hybrid Monte Carlo (HMC) scheme to improve samples generated by standard particle filters. Parallel marginalization is an efficient Markov chain Monte Carlo (MCMC) strategy that uses lower dimensional approximate marginal distributions of the target distribution to accelerate equilibration. As a validation the algorithm is tested on a 2516 dimensional, bimodal, stochastic model motivated by the Kuroshio current that runs along the Japanese coast. The results of this test indicate that the method is an attractive alternative for problems that require the generality of a particle filter but have been inaccessible due to the limitations of standard particle filtering strategies.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Inspection of next-generation EUV resists with nano-projectile secondary ion mass spectrometry

There is a rapidly growing need for new materials for extreme ultraviolet (EUV) lithography, which incorporate high EUV absorbing elements. Hybrid resists, which are a combination of inorganic and organic moieties, are promising as they offer both high EUV sensitivity and high etch resistance. However, there is a glaring lack of methods to examine the uniformity of these important materials at the nanoscale. We examine the capabilities of nano-projectile secondary ion mass spectrometry (NP-SIMS) to investigate hybrid resists. NP-SIMS is a mass spectrometry-based technique with high lateral resolution. Using NP-SIMS, a surface is probed stochastically with a suite of individual projectiles, 10 6 to 10 7 in total, separated in time and space. Examining these individual mass spectra allows for nanoscale investigation of the uniformity of the surface. We evaluated the performance of NP-SIMS using samples of poly(methyl methacrylate) (PMMA) infiltrated with varying amounts of InO x via vapor-phase infiltration (VPI), an organic–inorganic hybridization method derived from atomic layer deposition. Here, we found that NP-SIMS measurements contained abundant characteristic ions related to both the PMMA and infiltrated In. The intensity of In atomic ions increased linearly with the number of infiltration cycles; however, the uniformity of In and PMMA varied with the number of infiltration cycles. After one cycle, we found that both the PMMA and In were relatively inhomogeneous. The homogeneity improved with subsequent infiltration cycles. In addition, NP-SIMS measurements contained characteristic ions related to the infiltration reaction and provided insights into the VPI mechanism. Overall, the results show that NP-SIMS is capable of examining both the inorganic and organic moieties in a hybrid resist and will be an important method for understanding the performance of these materials for use in EUV lithography.

36 MATERIALS SCIENCE↗

Polynomial Chaos Surrogate Construction for Random Fields with Parametric Uncertainty

Engineering and applied science rely on computational experiments to rigorously study physical systems. The mathematical models used to probe these systems are highly complex, and sampling-intensive studies often require prohibitively many simulations for acceptable accuracy. Surrogate models provide a means of circumventing the high computational expense of sampling such complex models. In particular, polynomial chaos expansions (PCEs) have been successfully used for uncertainty quantification studies of deterministic models where the dominant source of uncertainty is parametric. We discuss an extension to conventional PCE surrogate modeling to enable surrogate construction for stochastic computational models that have intrinsic noise in addition to parametric uncertainty. We develop a PCE surrogate on a joint space of intrinsic and parametric uncertainty, enabled by Rosenblatt transformations, which are evaluated via kernel density estimation of the associated conditional cumulative distributions. Furthermore, we extend the construction to random field data via the Karhunen–Loève expansion. We then take advantage of closed-form solutions for computing PCE Sobol indices to perform a global sensitivity analysis of the model which quantifies the intrinsic noise contribution to the overall model output variance. Additionally, the resulting joint PCE is generative in the sense that it allows generating random realizations at any input parameter setting that are statistically approximately equivalent to realizations from the underlying stochastic model. The method is demonstrated on a chemical catalysis example model and a synthetic example controlled by a parameter that enables a switch from unimodal to bimodal response distributions.

97 MATHEMATICS AND COMPUTING↗

Generating moment matching scenarios using optimization techniques

An optimization based method is proposed to generate moment matching scenarios for numerical integration and its use in stochastic programming. The main advantage of the method is its flexibility: it can generate scenarios matching any prescribed set of moments of the underlying distribution rather than matching all moments up to a certain order, and the distribution can be defined over an arbitrary set. This allows for a reduction in the number of scenarios and allows the scenarios to be better tailored to the problem at hand. The method is based on a semi-infinite linear programming formulation of the problem that is shown to be solvable with polynomial iteration complexity. A practical column generation method is implemented. The column generation subproblems are polynomial optimization problems; however, they need not be solved to optimality. It is found that the columns in the column generation approach can be efficiently generated by random sampling. The number of scenarios generated matches a lower bound of Tchakaloff's. The rate of convergence of the approximation error is established for continuous integrands, and an improved bound is given for smooth integrands. Extensive numerical experiments are presented in which variants of the proposed method are compared to Monte Carlo and quasi-Monte Carlo methods on both numerical integration problems and stochastic optimization problems. The benefits of being able to match any prescribed set of moments, rather than all moments up to a certain order, is also demonstrated using optimization problems with 100-dimensional random vectors. Here, empirical results show that the proposed approach outperforms Monte Carlo and quasi-Monte Carlo based approaches on the tested problems.

97 MATHEMATICS AND COMPUTING↗

Generative Ensemble Regression: Learning Particle Dynamics from Observations of Ensembles with Physics-informed Deep Generative Models

Here, we propose a new method for inferring the governing stochastic ordinary differential equations (SODEs) by observing particle ensembles at discrete and sparse time instants, i.e., multiple “snapshots.” Particle coordinates at a single time instant, possibly noisy or truncated, are recorded in each snapshot but are unpaired across the snapshots. By training a physics-informed generative model that generates “fake” sample paths, we aim to fit the observed particle ensemble distributions with a curve in the probability measure space, which is induced from the inferred particle dynamics. We employ different metrics to quantify the differences between distributions, e.g., the sliced Wasserstein distances and the adversarial losses in generative adversarial networks. We refer to this method as generative “ensemble-regression” (GER), in analogy to the classic “point-regression,” where we infer the dynamics by performing regression in the Euclidean space. We illustrate the GER by learning the drift and diffusion terms of particle ensembles governed by SODEs with Brownian motions and Lévy processes up to 100 dimensions. We also discuss how to treat cases with noisy or truncated observations. Apart from systems consisting of independent particles, we also tackle nonlocal interacting particle systems with unknown interaction potential parameters by constructing a physics-informed loss function. Finally, we investigate scenarios of paired observations and discuss how to reduce the dimensionality in such cases by proving a convergence theorem that provides theoretical support.

97 MATHEMATICS AND COMPUTING↗

A hybrid method for hydrodynamic-kinetic flow Part I: A particle-grid method for reducing stochastic noise in kinetic regimes

In this work we present a hybrid particle-grid Monte Carlo method for the Boltzmann equation, which is characterized by a significant reduction of the stochastic noise in the kinetic regime. The hybrid method is based on a first order splitting in time to separate the transport from the relaxation step. The transport step is solved by a deterministic scheme, while a hybrid DSMC-based method is used to solve the collision step. Such a hybrid scheme is based on splitting the solution in a collisional and a non-collisional part at the beginning of the collision step, and the DSMC method is used to solve the relaxation step for the collisional part of the solution only. This is accomplished by sampling only the fraction of particles candidate for collisions from the collisional part of the solution, performing collisions as in a standard DSMC method, and then projecting the particles back onto a velocity grid to compute a piecewise constant reconstruction for the collisional part of the solution. The latter is added to a piecewise constant reconstruction of the non-collisional part of the solution, which in fact remains unchanged during the relaxation step. Numerical results show that the stochastic noise is significantly reduced at large Knudsen numbers with respect to the standard DSMC method. Indeed in this algorithm, the particle scheme is applied only on the collisional part of the solution, so only this fraction of the solution is affected by stochastic fluctuations. But since the collisional part of the solution reduces as the Knudsen number increases, stochastic noise reduces as well at large Knudsen numbers.

97 MATHEMATICS AND COMPUTING↗

Extending the computational reach of a superconducting qutrit processor

Quantum computing with qudits is an emerging approach that exploits a larger, more connected computational space, providing advantages for many applications, including quantum simulation and quantum error correction. Nonetheless, qudits are typically afflicted by more complex errors and suffer greater noise sensitivity which renders their scaling difficult. In this work, we introduce techniques to tailor arbitrary qudit Markovian noise to stochastic Weyl–Heisenberg channels and mitigate noise that commutes with our Clifford and universal two-qudit gate in generic qudit circuits. We experimentally demonstrate these methods on a superconducting transmon qutrit processor, and benchmark their effectiveness for multipartite qutrit entanglement and random circuit sampling, obtaining up to 3× improvement in our results. To the best of our knowledge, this constitutes the first-ever error mitigation experiment performed on qutrits. Our work shows that despite the intrinsic complexity of manipulating higher-dimensional quantum systems, noise tailoring and error mitigation can significantly extend the computational reach of today’s qudit processors.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Studying rare nonadiabatic dynamics with transition path sampling quantum jump trajectories

We present a method to study rare nonadiabatic dynamics in open quantum systems using transition path sampling and quantum jump trajectories. As with applications of transition path sampling to classical dynamics, the method does not rely on prior knowledge of transition states or reactive pathways and thus can provide mechanistic insight into ultrafast relaxation processes in addition to their associated rates. In particular, we formulate a quantum path ensemble using the stochastic realizations of an unravelled quantum master equation, which results in trajectories that can be conditioned on starting and ending in particular quantum states. Because the dynamics rigorously obeys detailed balance, rate constants can be evaluated from reversible work calculations in this conditioned ensemble, allowing for branching ratios and yields to be computed in an unbiased manner. We illustrate the utility of this method with three examples: energy transfer in a donor-bridge-acceptor model, and models of photo-induced proton-coupled electron transfer and thermally activated electron transfer. These examples demonstrate the efficacy of path ensemble methods and pave the way for their use in studying complex reactive quantum dynamics.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

An Online Approach to Solve the Dynamic Vehicle Routing Problem with Stochastic Trip Requests for Paratransit Services

Many transit agencies operating paratransit and microtransit services have to respond to trip requests that arrive in real-time, which entails solving hard combinatorial and sequential decision-making problems under uncertainty. To avoid decisions that lead to significant inefficiency in the long term, vehicles should be allocated to requests by optimizing a non-myopic utility function or by batching requests together and optimizing a myopic utility function. While the former approach is typically offline, the latter can be performed online. We point out two major issues with such approaches when applied to paratransit services in practice. First, it is difficult to batch paratransit requests together as they are temporally sparse. Second, the environment in which transit agencies operate changes dynamically (e.g., traffic conditions can change over time), causing the estimates that are learned offline to become stale. To address these challenges, we propose a fully online approach to solve the dynamic vehicle routing problem (DVRP) with time windows and stochastic trip requests that is robust to changing environmental dynamics by construction. We focus on scenarios where requests are relatively sparse—our problem is motivated by applications to paratransit services. We formulate DVRP as a Markov decision process and use Monte Carlo tree search to evaluate actions for any given state. Accounting for stochastic requests while optimizing a non-myopic utility function is computationally challenging; indeed, the action space for such a problem is intractably large in practice. To tackle the large action space, we leverage the structure of the problem to design heuristics that can sample promising actions for the tree search. Our experiments using real-world data from our partner agency show that the proposed approach outperforms existing state-of-the-art approaches both in terms of performance and robustness.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Frequent ultraviolet brightenings observed in a solar active region with solar maximum mission

Observations of the temporal behavior of ultraviolet emission from bright points within an active region of the sun are reported. Frequent and rapid brightenings in Si IV and O IV line emission are seen. The observations suggest that intermittent heating events of modest amplitude are occurring at many sites within an active region. By selecting the brightest site at any given time within an active region and then sampling its behavior in detail within a 120 s interval, it is found that about two-thirds of the samples show variations of the Si IV line intensity. The brightenings typically last about 40-60 s; intensity increases of about 20-100 percent are frequently observed. The results suggest that heating due to magnetic field reconnection within an active region is proceeding almost stochastically. Events involving only a modest release of energy occur the most frequently.

Porter, J. G.↗

Monte Carlo simulation of launchsite winds at Kennedy Space Center

This paper develops and validates an easily implemented model for simulating random horizontal wind profiles over the Kennedy Space Center (KSC) at Cape Canaveral, Florida. The model is intended for use in Monte Carlo launch vehicle simulations of the type employed in mission planning, where the large number of profiles needed for statistical fidelity of such simulation experiments makes the use of actual wind measurements impractical. The model is based on measurements made at KSC and represents vertical correlations by a decaying exponential model which is parameterized via least-squares parameter optimization against the sample data. The validity of the model is evaluated by comparing two Monte Carlo simulations of an asymmetric, heavy-lift launch vehicle. In the first simulation, the measured wind profiles are used, while in the second, the wind profiles are generated using the stochastic model. The simulations indicate that the use of either the measured or simulated wind field results in similar launch vehicle performance.

Queen, Eric M.↗

New bounding and decomposition approaches for MILP investment problems: Multi-area transmission and generation planning under policy constraints

A novel two-phase bounding and decomposition approach to compute optimal and near-optimal solutions to large-scale mixed-integer investment planning problems is proposed and it considers a large number of operating subproblems, each of which is a convex optimization. Our motivating application is the planning of power transmission and generation in which policy constraints are designed to incentivize high amounts of intermittent generation in electric power systems. The bounding phase exploits Jensen’s inequality to define a lower bound, which we extend to stochastic programs that use expected-value constraints to enforce policy objectives. The decomposition phase, in which the bounds are tightened, improves upon the standard Benders’ algorithm by accelerating the convergence of the bounds. The lower bound is tightened by using a Jensen’s inequality-based approach to introduce an auxiliary lower bound into the Benders master problem. Upper bounds for both phases are computed using a sub-sampling approach executed on a parallel computer system. Numerical results show that only the bounding phase is necessary if loose optimality gaps are acceptable. But, the decomposition phase is required to attain optimality gaps. Moreover, use of both phases performs better, in terms of convergence speed, than attempting to solve the problem using just the bounding phase or regular Benders decomposition separately.

97 MATHEMATICS AND COMPUTING↗

Calculating uncertainty on K-effective with MONK10 - 14583

Criticality safety assessments require a demonstration that a particular configuration of fissile material has an adequate sub-critical margin (k-effective sufficiently below unity) to ensure that the risk of criticality under normal operation and accident conditions is acceptable. The required sub-critical margin depends upon the uncertainty in the estimated value of k-effective. The uncertainty in the calculated value of k-effective arises from a number of sources, including: manufacturing tolerances on input data to the code (affecting geometry, compositions and densities), uncertainty in the nuclear data used by the code, stochastic uncertainty resulting from Monte Carlo simulation and modelling approximations/errors, including the inevitable bugs in the software. The ANSWERS Software Service, in collaboration with industrial partners, is developing a number of techniques to better understand and quantify uncertainty on predicted values of k-effective, using MONK. The SPRUCE utility code has been developed to allow uncertainty to be estimated using sampling methods. This can include the sampling of input parameters (including dimensions, compositions and densities) from statistical distributions. It can also include sampling different nuclear data libraries. A set of nuclear data libraries has been generated for this purpose by sampling from statistical distributions that represent the uncertainties in the published nuclear data evaluated files; a set of libraries has been produced for Latin Hypercube Sampling. By varying the input data and nuclear data, separate and combined uncertainties due to manufacturing tolerances and nuclear data can be derived. By performing least squares fitting on the results it is also possible to estimate the contribution of each of the uncertain inputs and a sensitivity method in MONK can break down the nuclear data uncertainty. (authors)

42 ENGINEERING↗

A Predictive Prescription Framework for Stochastic Unit Commitment Using Boosting Ensemble Learning Algorithms

To take unit commitment (UC) decisions under uncertain load, most existing stochastic optimization (SO) frameworks adopt a generic representation of uncertainty. While load levels that materialize on a particular day are influenced by various covariates (such as the day of the week or temperature), SO frameworks typically disregard such side observations, wasting actionable information that could significantly enhance decision quality. Here, this article proposes a contextual SO (CSO) framework for UC under uncertain load, which can effectively exploit covariate observations in conjunction with a class of machine learning (ML) algorithms to improve the out-of-sample performance of UC decisions. It shows how three ML algorithms, adaptive boosting, gradient boosted trees, and extreme gradient boosting, can be used to this end, constituting the first application of these algorithms in any CSO framework. Using real-world data harvested from the New York ISO grid, we measure the out-of-sample performance of the framework in terms of total operation cost, shed load values, locational marginal prices, and total payments by the loads, against several benchmark methods proposed in the literature. The article has an online companion (Yurdakul et al.), wherein we present additional results and lay out further mathematical formulations used in this work.

42 ENGINEERING↗