Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “BAYESIAN STATISTICS”

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 667 records · Page 37

An Efficient Bayesian Approach to Learning Droplet Collision Kernels: Proof of Concept Using “Cloudy,” a New n -Moment Bulk Microphysics Scheme

The small-scale microphysical processes governing the formation of precipitation particles cannot be resolved explicitly by cloud resolving and climate models. Instead, they are represented by microphysics schemes that are based on a combination of theoretical knowledge, statistical assumptions, and fitting to data (“tuning”). Historically, tuning was done in an ad hoc fashion, leading to parameter choices that are not explainable or repeatable. Recent work has treated it as an inverse problem that can be solved by Bayesian inference. The posterior distribution of the parameters given the data—the solution of Bayesian inference—is found through computationally expensive sampling methods, which require over $\mathcal{O}$(10 5 ) evaluations of the forward model; this is prohibitive for many models. We present a proof of concept of Bayesian learning applied to a new bulk microphysics scheme named “Cloudy,” using the recently developed Calibrate-Emulate-Sample (CES) algorithm. Cloudy models collision-coalescence and collisional breakup of cloud droplets with an adjustable number of prognostic moments and with easily modifiable assumptions for the cloud droplet mass distribution and the collision kernel. The CES algorithm uses machine learning tools to accelerate Bayesian inference by reducing the number of forward evaluations needed to $\mathcal{O}$(10 2 ). It also exhibits a smoothing effect when forward evaluations are polluted by noise. In a suite of perfect-model experiments, we show that CES enables computationally efficient Bayesian inference of parameters in Cloudy from noisy observations of moments of the droplet mass distribution. In an additional imperfect-model experiment, a collision kernel parameter is successfully learned from output generated by a Lagrangian particle-based microphysics model.

54 ENVIRONMENTAL SCIENCES↗

On-the-fly autonomous control of neutron diffraction via physics-informed Bayesian active learning

We demonstrate the first live, autonomous control over neutron diffraction experiments by developing and deploying ANDiE: the autonomous neutron diffraction explorer. Neutron scattering is a unique and versatile characterization technique for probing the magnetic structure and behavior of materials. However, instruments at neutron scattering facilities in the world is limited, and instruments at such facilities are perennially oversubscribed. We demonstrate a significant reduction in experimental time required for neutron diffraction experiments by implementation of autonomous navigation of measurement parameter space through machine learning. Prior scientific knowledge and Bayesian active learning are used to dynamically steer the sequence of measurements. We show that ANDiE can experimentally determine the magnetic ordering transition of both MnO and Fe 1.09 Te all while providing a fivefold enhancement in measurement efficiency. Furthermore, in a hypothesis testing post-processing step, ANDiE can determine transition behavior from a set of possible physical models. ANDiE's active learning approach is broadly applicable to a variety of neutron-based experiments and can open the door for neutron scattering as a tool of accelerated materials discovery.

36 MATERIALS SCIENCE↗

Metalearners for estimating heterogeneous treatment effects using machine learning

There is growing interest in estimating and analyzing heterogeneous treatment effects in experimental and observational studies. We describe a number of metaalgorithms that can take advantage of any supervised learning or regression method in machine learning and statistics to estimate the conditional average treatment effect (CATE) function. Metaalgorithms build on base algorithms—such as random forests (RFs), Bayesian additive regression trees (BARTs), or neural networks—to estimate the CATE, a function that the base algorithms are not designed to estimate directly. We introduce a metaalgorithm, the X-learner, that is provably efficient when the number of units in one treatment group is much larger than in the other and can exploit structural properties of the CATE function. For example, if the CATE function is linear and the response functions in treatment and control are Lipschitz-continuous, the X-learner can still achieve the parametric rate under regularity conditions. We then introduce versions of the X-learner that use RF and BART as base learners. In extensive simulation studies, the X-learner performs favorably, although none of the metalearners is uniformly the best. In two persuasion field experiments from political science, we demonstrate how our X-learner can be used to target treatment regimes and to shed light on underlying mechanisms. A software package is provided that implements our methods.

42 ENGINEERING↗

GAMMA-RAY BURST FLARES: ULTRAVIOLET/OPTICAL FLARING. I

We present a previously unused method for the detection of flares in gamma-ray burst (GRB) light curves and use this method to detect flares in the ultraviolet/optical. The algorithm makes use of the Bayesian Information Criterion to analyze the residuals of the fitted light curve, removing all major features, and to determine the statistically best fit to the data by iteratively adding additional ''breaks'' to the light curve. These additional breaks represent the individual components of the detected flares: T{sub start}, T{sub stop}, and T{sub peak}. We present the detection of 119 unique flaring periods detected by applying this algorithm to light curves taken from the Second Swift Ultraviolet/Optical Telescope (UVOT) GRB Afterglow Catalog. We analyzed 201 UVOT GRB light curves and found episodes of flaring in 68 of the light curves. For those light curves with flares, we find an average number of {approx}2 flares per GRB. Flaring is generally restricted to the first 1000 s of the afterglow, but can be observed and detected beyond 10{sup 5} s. More than 80% of the flares detected are short in duration with {Delta}t/t of <0.5. Flares were observed with flux ratios relative to the underlying light curve of between 0.04 and 55.42. Many of the strongest flares were also seen at greater than 1000 s after the burst.

79 ASTRONOMY AND ASTROPHYSICS↗

A novel application of cosmic rays for spent nuclear fuel monitoring - 15738

Monitoring nuclear waste and controlling spent nuclear fuel at its source is one of the main strategies to minimize the risks of nuclear proliferation and reduce potential homeland threats. Storage of high level waste from a fission reactor typically involves placing spent fuel in sealed dry casks. After the spent nuclear fuel has been placed inside the dry cask, the cask is welded, not allowing for visual inspection. Conventional methods for examining the interior of materials e.g., X-rays, are limited by the fact that they cannot penetrate very dense well-shielded objects. Instead, cosmic ray muons have the potential to allow for assessment of spent nuclear fuel stored within sealed dense dry casks. Objectives: 1.Exploit the ability of cosmic ray muons to penetrate high density materials allowing the distribution of materials within the object to be inferred from the muon tracks. 2. Develop a monitoring system that tracks the path of cosmic ray muons, including their deflection and attenuation, to identify the contents of a closed dry cask. Methods: - Monte-Carlo simulations using Geant4; - Bayesian decision theory for signal classification; - Wavelet analysis for noise cancellation; - Chaotic Neural Networks for signal encryption. Results: 1. Preliminary statistical analysis proved that the method is feasible and has the ability to identify the presence of spent nuclear fuel in a dry cask. 2. Using monoenergetic muons, a dry cask with one fuel assembly missing can be correctly identified with error rate <1%. Broader merit: Once completed, the proposed project for cosmic ray muon transport will enable a new and potentially transformative approach towards efficient, inexpensive and remote cosmic ray muon applications in archaeology, medicine and safeguarding of nuclear materials.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Hierarchical Gaussian Random Field Sampling for Multilevel Markov Chain Monte Carlo: Coupling Stochastic Partial Differential Equation and the Karhunen–Loève Decomposition

This work introduces structure preserving hierarchical decompositions for sampling Gaussian random fields (GRFs) within the context of multilevel Bayesian inference in high-dimensional space. Existing scalable hierarchical sampling methods, such as those based on stochastic partial differential equations (SPDEs), often reduce the dimensionality of the sample space at the cost of accuracy of inference. Other approaches, such that those based on Karhunen-Loève (KL) expansions, offer sample space dimensionality reduction but sacrifice GRF representation accuracy and ergodicity of the Markov chain Monte Carlo (MCMC) sampler and are computationally expensive for high-dimensional problems. The proposed method integrates the dimensionality reduction capabilities of KL expansions with the scalability of SPDE-based sampling, thereby providing a robust, unified framework for high-dimensional uncertainty quantification (UQ) that is scalable and accurate, preserves ergodicity, and offers dimensionality reduction of the sample space. The hierarchy in our multilevel algorithm is derived from the geometric multigrid hierarchy. By constructing a hierarchical decomposition that maintains the covariance structure across the levels in the hierarchy, the approach enables efficient coarse-to-fine sampling while ensuring that all samples are drawn from the desired distribution. The effectiveness of the proposed method is demonstrated on a benchmark subsurface flow problem, demonstrating its effectiveness in improving computational efficiency and statistical accuracy. Furthermore, our proposed technique is more efficient and accurate and displays better convergence properties than existing methods for high-dimensional Bayesian inference problems.

Gaussian random fields↗

Scale Mixtures of Frechet Distributions as Asymptotic Approximations of Extreme Precipitation

This paper is a further development of the results of [20] where, based on the negative binomial model for the duration of wet periods measured in days [16], an asymptotic approximation was proposed for the distribution of the maximum daily precipitation volume within a wet period. This approximation has the form of a scale mixture of the Fr´echet distribution with the gamma mixing distribution and coincides with the distribution of a positive power of a random variable having the Snedecor–Fisher distribution. Here we extend this result to the mth extremes, m ∈ ℕ, and sample quantiles. The proof of this result is based on the representation of the negative binomial distribution as a mixed geometric (and hence, mixed Poisson) distribution [17] and limit theorems for extreme order statistics in samples with random sizes having mixed Poisson distributions [10]. Some analytic properties of the obtained limit distribution are described. In particular, it is demonstrated that under certain conditions the limit distribution of the maximum precipitation is mixed exponential and hence, is infinitely divisible. It is shown that under the same conditions this limit distribution can be represented as a scale mixture of stable or Weibull or Pareto or folded normal laws. The corresponding product representations for the limit random variable can be used for its computer simulation. The results of fitting this distribution to real data are presented illustrating high adequacy of the proposed model. It is also shown that the limit distribution of sample quantiles is the well-known Student distribution. Several methods are proposed for the estimation of the parameters of the asymptotic distributions. The obtained mixture representations for the limit laws and the corresponding asymptotic approximations provide better insight into the nature of mixed probability (“Bayesian”) models.

97 MATHEMATICS AND COMPUTING↗

How data science methods can improve the quality and efficiency of ICF and HEDP research

Data Science methods (many that are Bayesian based) are widely used in the physical sciences to estimate model parameters from experimental data, synthesize heterogeneous data, calibrate models, design experiments, and determine statistical significance of data. These methods provide a wealth of advantages over traditional analysis techniques because: 1) uncertainties are rigorously defined and propagated naturally through complex systems including covariance, 2) prior information is captured within the analysis framework (including rad-MHD and rad-hydro simulations), 3) competing models can be selected and/or ruled out using quantitative criteria, and 4) complex, heterogeneous data can be incorporated simultaneously. While these methods have been widely adopted as the gold standard in fields such as particle physics, astronomy, and biology, they have been slow to catch on in Inertial Confinement Fusion (ICF) and High Energy Density Physics (HEDP) research. Recently, several teams at LLNL, SNL, LANL, and the LLE have been exploring the use of these tools in their research and have found success. Here we propose that a concerted effort to consolidate these independent research efforts by developing and deploying common tools for use across the complex can revolutionize the way we approach data analysis, assimilation of theory and experiment, and decision making. The Bayesian formalism provides a means to accomplish this, but we are lacking certain infrastructure to make it happen on a large scale. Furthermore, once adopted, these techniques can be used to develop standards by which discoveries can be judged, similar to the so-called 5σ rule in high energy particle physics. Such standards may be used in the future to address the issue of unknown reproducibility in ICF and HED experiments caused by low shot rate and high cost per experiment. Our goals as a group are to advance the state of the art in HED measurement science by enabling: 1) better inferences from data with well-defined uncertainties, 2) better use of the data we have and continue to collect, 3) intelligent synthesis of data, 4) evaluation of the statistical significance of our data, and 5) informed decision making regarding the design of new experiments and instruments.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

A Bayesian Integration Model of High-Throughput Proteomics and Metabolomics Data for Improved Early Detection of Microbial Infections

High-throughput (HTP) technologies offer the capability to evaluate the genome, proteome, and metabolome of organisms at a global scale. This opens up new opportunities to define complex signatures of disease that involve signals from multiple types of biomolecules. Integrating these data types however is difficult due to the heterogeneity of the data. We present a Bayesian approach to integration that uses posterior probabilities to assign class memberships to samples using individual and multiple data sources; these probabilities are based on lower level likelihood functions derived from standard statistical learning algorithms. We demonstrate this approach on microbial infections of mice, where the bronchial alveolar lavage fluid was analyzed by two HTP proteomic and one HTP metabolomic technologies. We demonstrate that integration of the three datasets improves classification accuracy to ~89% from the best individual dataset at ~83%. In addition, we present a new visualization tool called Visual Integration for Bayesian Evaluation (VIBE) that allows the user to observe classification accuracies at the class level and evaluate classification accuracies on any subset of available data types based on the posterior probability models defined for the individual and integrated data.

59 BASIC BIOLOGICAL SCIENCES↗

A Bayesian Framework for Spectral Reprojection

Abstract Fourier partial sum approximations yield exponential accuracy for smooth and periodic functions, but produce the infamous Gibbs phenomenon for non-periodic ones. Spectral reprojection resolves the Gibbs phenomenon by projecting the Fourier partial sum onto a Gibbs complementary basis, often prescribed as the Gegenbauer polynomials. Noise in the Fourier data and the Runge phenomenon both degrade the quality of the Gegenbauer reconstruction solution, however. Motivated by its theoretical convergence properties, this paper proposes a new Bayesian framework for spectral reprojection, which allows a greater understanding of the impact of noise on the reprojection method from a statistical point of view. We are also able to improve the robustness with respect to the Gegenbauer polynomials parameters. Finally, the framework provides a mechanism to quantify the uncertainty of the solution estimate.

Li, Tongtong (ORCID:0000000276644764)↗

Visible core spectroscopy at Wendelstein 7-X

This paper presents an overview of recent hardware extensions and data analysis developments to the Wendelstein 7-X visible core spectroscopy systems. These include upgrades to prepare the in-vessel components for long-pulse operation, nine additional spectrometers, a new line of sight array for passive spectroscopy, and a coherence imaging charge exchange spectroscopy diagnostic. Progress in data analysis includes ion temperatures and densities from multiple impurity species, a statistical comparison with x-ray crystal spectrometer measurements, neutral density measurements from thermal passive Balmer-alpha emission, and a Bayesian analysis of active hydrogen emission, which is able to infer electron density and main ion temperature profiles.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Non-Intrusive Appliance Identification with Appliance-Specific Networks

The problem of noninstrusive load monitoring (NILM) is usually formulated as a single-channel blind source separation task, whose successful solution enable fast and convenient load identification and energy disaggregation. When applied at test time, NILM algorithms aim to identify the operating characteristics of individual appliances from an aggregate power measurement of the entire house. Recent advances in deep learning gave rise to many methods that mostly focus on learning a direct mapping from aggregate measurement to individual appliance power. However, these methods are not only computationally expensive, but they often suffer from overfitting and do not generalize very well. In this article, we propose a novel NILM method that leverages advances in statistical learning that have not been properly applied in this domain before. The proposed method consists of three stages: first, a Bayesian nonparametric learning-based approach for appliance state extraction; second, synthetic minority oversampling technique for data augmentation and mitigating the heavy imbalance in switching events; and third, appliance-specific lightweight long short-term memory networks for status classification for each appliance. Here, we adopt a “differential” input (the difference before and after the switching event) to reduce the complexity of network training and make the proposed method robust to multiappliance switching events. Experiments are conducted to demonstrate the effectiveness of the proposed method, achieving superior performance when compared to recent methods. An ablation study is conducted to demonstrate the effectiveness of each module of our method. Finally, we investigate the quality of generated synthetic samples.

42 ENGINEERING↗

Uncertainty Management for Diagnostics and Prognostics of Batteries using Bayesian Techniques

Uncertainty management has always been the key hurdle faced by diagnostics and prognostics algorithms. A Bayesian treatment of this problem provides an elegant and theoretically sound approach to the modern Condition- Based Maintenance (CBM)/Prognostic Health Management (PHM) paradigm. The application of the Bayesian techniques to regression and classification in the form of Relevance Vector Machine (RVM), and to state estimation as in Particle Filters (PF), provides a powerful tool to integrate the diagnosis and prognosis of battery health. The RVM, which is a Bayesian treatment of the Support Vector Machine (SVM), is used for model identification, while the PF framework uses the learnt model, statistical estimates of noise and anticipated operational conditions to provide estimates of remaining useful life (RUL) in the form of a probability density function (PDF). This type of prognostics generates a significant value addition to the management of any operation involving electrical systems.

Saha, Bhaskar↗

Uncertainty Estimates of Psychoacoustic Thresholds Obtained from Group Tests

Adaptive psychoacoustic test methods, in which the next signal level depends on the response to the previous signal, are the most efficient for determining psychoacoustic thresholds of individual subjects. In many tests conducted in the NASA psychoacoustic labs, the goal is to determine thresholds representative of the general population. To do this economically, non-adaptive testing methods are used in which three or four subjects are tested at the same time with predetermined signal levels. This approach requires us to identify techniques for assessing the uncertainty in resulting group-average psychoacoustic thresholds. In this presentation we examine the Delta Method of frequentist statistics, the Generalized Linear Model (GLM), the Nonparametric Bootstrap, a frequentist method, and Markov Chain Monte Carlo Posterior Estimation and a Bayesian approach. Each technique is exercised on a manufactured, theoretical dataset and then on datasets from two psychoacoustics facilities at NASA. The Delta Method is the simplest to implement and accurate for the cases studied. The GLM is found to be the least robust, and the Bootstrap takes the longest to calculate. The Bayesian Posterior Estimate is the most versatile technique examined because it allows the inclusion of prior information.

Rathsam, Jonathan↗

Confidence Intervals for Laboratory Sonic Boom Annoyance Tests

Commercial supersonic flight is currently forbidden over land because sonic booms have historically caused unacceptable annoyance levels in overflown communities. NASA is providing data and expertise to noise regulators as they consider relaxing the ban for future quiet supersonic aircraft. One deliverable NASA will provide is a predictive model for indoor annoyance to aid in setting an acceptable quiet sonic boom threshold. A laboratory study was conducted to determine how indoor vibrations caused by sonic booms affect annoyance judgments. The test method required finding the point of subjective equality (PSE) between sonic boom signals that cause vibrations and signals not causing vibrations played at various amplitudes. This presentation focuses on a few statistical techniques for estimating the interval around the PSE. The techniques examined are the Delta Method, Parametric and Nonparametric Bootstrapping, and Bayesian Posterior Estimation.

Rathsam, Jonathan↗

Bayesian Calibration of the Community Land Model using Surrogates

We present results from the Bayesian calibration of hydrological parameters of the Community Land Model (CLM), which is often used in climate simulations and Earth system models. A statistical inverse problem is formulated for three hydrological parameters, conditioned on observations of latent heat surface fluxes over 48 months. Our calibration method uses polynomial and Gaussian process surrogates of the CLM, and solves the parameter estimation problem using a Markov chain Monte Carlo sampler. Posterior probability densities for the parameters are developed for two sites with different soil and vegetation covers. Our method also allows us to examine the structural error in CLM under two error models. We find that accurate surrogate models can be created for CLM in most cases. The posterior distributions lead to better prediction than the default parameter values in CLM. Climatologically averaging the observations does not modify the parameters’ distributions significantly. The structural error model reveals a correlation time-scale which can potentially be used to identify physical processes that could be contributing to it. While the calibrated CLM has a higher predictive skill, the calibration is under-dispersive.

Bayesian calibration↗