A Multi-Timescale Data-Driven Approach to Enhance Distribution System Observability
Not provided.
SEARCH · Search NASA
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.
Not provided.
Abstract Terrestrial gross primary production (GPP) represents the magnitude of CO 2 uptake through vegetation photosynthesis, and is a key variable for carbon cycles between the biosphere and atmosphere. Light use efficiency (LUE) models have been widely used to estimate GPP for its physiological mechanisms and availability of data acquisition and implementation, yet each individual GPP model has exhibited large uncertainties due to input errors and model structure, and further studies of systematic validation, comparison, and fusion of those models with eddy covariance (EC) site data across diverse ecosystem types are still needed in order to further improve GPP estimation. We here compared and fused five GPP models (VPM, EC‐LUE, GOL‐PEM, CHJ, and C‐Fix) across eight ecosystems based on FLUXNET2015 data set using the ensemble methods of Bayesian Model Averaging (BMA), Support Vector Machine (SVM), and Random Forest (RF) separately. Our results showed that for individual models, EC‐LUE gave a better performance to capture interannual variability of GPP than other models, followed by VPM and GLO‐PEM, while CHJ and C‐Fix were more limited in their estimation performance. We found RF and SVM were superior to BMA on merging individual models at various plant functional types (PFTs) and at the scale of individual sites. On the basis of individual models, the fusion methods of BMA, SVM, and RF were examined by a five‐fold cross validation for each ecosystem type, and each method successfully improved the average accuracy of estimation by 8%, 18%, and 19%, respectively.
This study present a test case in which the Bayesian Optimization method is applied to a simulation-based reactor core design optimization problem. The test case aims to showcase the potential of an automated design optimization algorithm for reactor designs by streamlining the reactor core design workflow, given the high computational cost of simulations. The contributions of this work are threefold. First, the existing HTGR model is converted into a simulation-based design optimization test case by developing a pipeline that enables modification of key design parameters and evaluates design performance based on simulation outputs. Second, Bayesian Optimization is implemented and adapted to demonstrate the feasibility of automatic design optimization for nuclear reactor core. Proposed approach leverages Gaussian Process models to characterize the relationship between design variables and performance metrics, while incorporating novel acquisition functions that balance exploration of the design space with exploitation of promising configurations. This implementation lays the foundation for the future developments of reactor design optimization algorithms.
Abstract not provided
High-resolution spectroscopic surveys of the Milky Way have entered the Big Data regime and have opened avenues for solving outstanding questions in Galactic archeology. However, exploiting their full potential is limited by complex systematics, whose characterization has not received much attention in modern spectroscopic analyses. In this work, we present a novel method to disentangle the component of spectral data space intrinsic to the stars from that due to systematics. Using functional principal component analysis on a sample of 18,933 giant spectra from APOGEE, we find that the intrinsic structure above the level of observational uncertainties requires ≈10 functional principal components (FPCs). Our FPCs can reduce the dimensionality of spectra, remove systematics, and impute masked wavelengths, thereby enabling accurate studies of stellar populations. To demonstrate the applicability of our FPCs, we use them to infer stellar parameters and abundances of 28 giants in the open cluster M67. We employ Sequential Neural Likelihood, a simulation-based Bayesian inference method that learns likelihood functions using neural density estimators, to incorporate non-Gaussian effects in spectral likelihoods. By hierarchically combining the inferred abundances, we limit the spread of the following elements in M67: Fe ≲ 0.02 dex; C ≲ 0.03 dex; O, Mg, Si, Ni ≲ 0.04 dex; Ca ≲ 0.05 dex; N, Al ≲ 0.07 dex (at 68% confidence). Our constraints suggest a lack of self-pollution by core-collapse supernovae in M67, which has promising implications for the future of chemical tagging to understand the star formation history and dynamical evolution of the Milky Way.
We construct efficient emulators for the computation of the infinite nuclear matter equation of state. These emulators are based on the subspace-projected coupled-cluster method for which we here develop a new algorithm called small-batch voting to eliminate spurious states that might appear when emulating quantum many-body methods based on a non-Hermitian Hamiltonian. The efficiency and accuracy of these emulators facilitate a rigorous statistical analysis within which we explore nuclear matter predictions for > 10 6 different parametrizations of a chiral interaction model with explicit Δ -isobars at next-to-next-to leading order. Constrained by nucleon-nucleon scattering phase shifts and bound-state observables of light nuclei up to He 4 , we use history matching to identify nonimplausible domains for the low-energy coupling constants of the chiral interaction. Within these domains we perform a Bayesian analysis using sampling and importance resampling with different likelihood calibrations and study correlations between interaction parameters, calibration observables in light nuclei, and nuclear matter saturation properties. Published by the American Physical Society 2024
Pilot plant test campaigns can be expensive and time-consuming. Therefore, it is of interest to maximize the amount of learning and the efficiency of the test campaign given the limited number of experiments that can be conducted. This work investigates the use of sequential design of experiments (SDOE) to overcome these challenges by demonstrating its usefulness for a recent solvent-based CO2 capture plant test campaign. Unlike traditional design of experiments methods, SDOE regularly uses information from ongoing experiments to determine the optimum locations in the design space for subsequent runs within the same experiment. However, there are challenges that need to be addressed, including reducing the high computational burden to efficiently update the model, and the need to incorporate the methodology into a computational tool. We address these challenges by applying SDOE in combination with a software tool, the Framework for Optimization, Quantification of Uncertainty and Surrogates (FOQUS) (Miller et al., 2014a, 2016, 2017). The results of applying SDOE on a pilot plant test campaign for CO2 capture suggests that relative to traditional design of experiments methods, SDOE can more effectively reduce the uncertainty of the model, thus decreasing technical risk. Future work includes integrating SDOE into FOQUS and using SDOE to support additional large-scale pilot plant test campaigns.
There is a nearly ubiquitous assumption in PSA that parameter values are at least piecewise-constant in time. As a result, Bayesian inference tends to incorporate many years of plant operation, over which there have been significant changes in plant operational and maintenance practices, plant management, etc. These changes can cause significant changes in parameter values over time; however, failure to perform Bayesian inference in the proper time-dependent framework can mask these changes. Failure to question the assumption of constant parameter values, and failure to perform Bayesian inference in the proper time-dependent framework were noted as important issues in NUREG/CR-6813, performed for the U. S. Nuclear Regulatory Commission’s Advisory Committee on Reactor Safeguards in 2003. That report noted that “industry lacks tools to perform time-trend analysis with Bayesian updating.” This paper describes an application of time-dependent Bayesian inference methods developed for the European Commission Ageing PSA Network. These methods utilize open-source software, implementing Markov chain Monte Carlo sampling. The paper also illustrates the development of a generic prior distribution, which incorporates multiple sources of generic data via weighting factors that address differences in key influences, such as vendor, component boundaries, conditions of the operating environment, etc.
The TENDL library is now established as one of the major nuclear data libraries in the world, striving for completeness and quality of nuclear data files for all isotopes, evaluation methods, processing and applied performance. To reach this status, some basic principles have been applied which sets it apart from other libraries: reproducible dedicated evaluations when differential data are available, through determination of nuclear models implemented in TALYS and their parameters, completeness (with or without experimental data), format and processing standardization, automation of production and reproducibility. In this paper, we will outline how such an approach has become a reality, and recall some of the past successes since the first TENDL release in 2008. Next, we will demonstrate the performance of the latest TENDL releases for different application fields, as well as new approaches for uncertainty quantification based on Bayesian inference methods and possible differential and integral adjustments. Also, current limitations of the library performances due to modelling and needs for new and more precise experimental data will be outlined.
The intense radiation emitted by luminous quasars dramatically alters the ionization state of their surrounding IGM. This so-called proximity effect extends out to tens of Mpc, and manifests as large coherent regions of enhanced Ly (Ly) forest transmission in absorption spectra of background sightlines. In this work, we present a novel method based on Ly forest tomography, which is capable of mapping these quasar "light echoes" in three dimensions. Using a dense grid (10-100) of faint (m r ≈ 24.7 mag) background galaxies as absorption probes, one can measure the ionization state of the IGM in the vicinity of a foreground quasar, yielding detailed information about the quasar's radiative history and emission geometry. An end-to-end analysis-combining cosmological hydrodynamical simulations post-processed with a quasar emission model, realistic estimates of galaxy number densities, and instrument + telescope throughput-is conducted to explore the feasibility of detecting quasar light echoes. We present a new, fully Bayesian statistical method that allows one to reconstruct quasar light echoes from thousands of individual low-S/N transmission measurements. Armed with this tool, we undertake an exhaustive parameter study and show that light echoes can be convincingly detected for luminous (M 1450 <-27.5 mag, corresponding to m 1450 < 18.4 mag at z≃3.6) quasars at redshifts 3 < z QSO < 5, and that a relative precision better than 20% on the quasar age can be achieved for individual objects in the expected range of ages between 1 and 100 Myr. The observational requirements are relatively modest: Moderate-resolution (R ≳ 750), multiobject spectroscopy at a low signal-to-noise ratio (S/N > 5) is sufficient, requiring three-hour integrations using existing instruments on 8 m class telescopes.
INTRODUCTION: Post-flight postural ataxia reflects adaptive changes to vestibulo-spinal reflexes and control strategies adopted for movement in weightlessness. Quantitative measures obtained during computerized dynamic posturography (CDP) from US and Russian programs provide insight into the effect of spaceflight duration in terms of both the initial decrements and recovery of postural stability. METHODS: CDP was obtained on 117 crewmembers following Shuttle flights lasting 4-17 days, and on 64 crewmembers following long-duration missions lasting 48-380 days. Although the number and timing of sessions varied, the goal was to characterize postural recovery pooling similar measures from different research and flight medicine programs. This report focuses on eyes closed, head erect conditions with either a fixed or sway-referenced base of support. A smaller subset of subjects repeated the sway-referenced condition while making pitch head movements (+/- 20deg at 0.33Hz). Equilibrium scores were derived from peak-to-peak anterior-posterior sway. Fall probability was modeled using Bayesian statistical methods to estimate parameters of a logit function. RESULTS: The standard Romberg condition was the least sensitive. Longer duration flights led to larger decrements in stability with sway-reference support during the first 1-2 days, although the timecourse of recovery was similar across flight duration with head erect. Head movements led to increased incidence of falls during the first week, with a significantly longer recovery following long duration flights. CONCLUSIONS: The diagnostic assessment of postural instability, and differences in the timecourse of postural recovery between short and long flight durations, are more pronounced during unstable support conditions requiring active head movements.
Applications of reduced basis method emulators are increasing in low-energy nuclear physics because they enable fast and accurate sampling of high-fidelity calculations, enabling robust uncertainty quantification. Here, in this paper, we develop, implement, and test two model-driven emulators based on the (Petrov-)Galerkin projection using the prototypical test case of two-body scattering with the Minnesota potential and a more realistic local chiral potential. The high-fidelity scattering equations are solved with the matrix Numerov method, a reformulation of the popular Numerov recurrence relation for solving special second-order differential equations as a linear system of coupled equations. A novel error estimator based on reduced-space residuals is applied to an active learning approach (a greedy algorithm) to choosing training samples (“snapshots”) for the emulator and contrasted with a proper orthogonal decomposition (POD) approach. Both approaches allow for computationally efficient offline-online decompositions, but the greedy approach requires many fewer snapshot calculations. These developments set the groundwork for emulating scattering observables based on chiral nucleon-nucleon and three-nucleon interactions and optical models, where computational speed-ups are necessary for Bayesian uncertainty quantification. Our emulators and error estimators are widely applicable to linear systems.
Limits on the Higgs production cross section at the Tevatron were placed using data with an integrated luminosity of 2.4 fb -1 from CDF. Limits over a Higgs mass range between 110 GeV and 200 GeV were determined, by calculating a limit at ten mass points distributed over this region. The analysis exclusively searches for Higgs produced by top-quark mediated gluon fusion and then decaying into two W bosons. Only leptonic decay channels of the W are considered, such that the final event signature consists of ee, eµ, or µµ with missing energy from undetected neutrinos. After an evaluation of alternative techniques, a neural net was selected as the best method for increasing the sensitivity of the measurement. The BFGS neural net training technique was selected as the most efficient method. A Bayesian Likelihood technique was used to place limits on the observed Higgs production cross section, and an expected limit was calculated by running 10,000 pseudo experiments. The 160 GeV mass point was the most most sensitive, achieving an expected limit 4.1 times the Standard Model prediction cross section at a 95% Confidence Level. Observed limits are within 1 σ of the expected limit below a mass point of ii 160 GeV. Above this, observed limits are higher than the expected limits, within 2 σ. The lowest observed limit was also at MH =160 GeV with a limit of 6.85 times the Standard Model prediction at a 95% Confidence Level. A new method for increasing the sensitivity of the measurement was proposed and investigated, but unused in the analysis.
Climate models often exhibit biases in their precipitation predictions, particularly underestimating high-intensity events and overestimating low precipitation. Deep learning approaches offer promising solutions, but their epistemic uncertainty associated with a deep learning–based bias correction method has not previously been quantified for reliable downstream climate impact studies. While methods for capturing the epistemic uncertainty in deep learning frameworks exist, there is currently no consensus on the best method. In this work, we compare three uncertainty quantification (UQ) methods—Deep Ensembles (DEns), Monte Carlo Dropout (MCD), and Flipout—by assessing the reliability of their uncertainty estimates using standard measures such as sharpness and calibration. These UQ methods are applied to an existing deep learning precipitation bias correction model known as UFNet: a coupled U-Net and fully connected neural network. The methods utilized to assess the models’ uncertainties are 1) calibration, which ensures that the expected probabilities of the model align with reality and 2) sharpness, which is a measure of the precision of the model’s probabilistic predictions. Of the three UQ methods evaluated, the DEns and MCD methods demonstrated the best-calibrated performance (expected calibration error of 0.36 and 0.35, respectively), compared to Flipout (0.58). In contrast, Flipout had the sharpest predictions and the highest metric performance in bias correcting precipitation—especially for higher-order moments such as kurtosis with a spatial correlation of 72% compared to 32% and 55% spatial correlation for DEns and MCD, respectively. Of the three UQ methods, MCD was found to be the most suitable method for UQ purposes based on its calibration, sharpness, and computational requirements.
Here, we report on methods and results for a model-based flood hazard assessment we have conducted for the Hudson River from New York City to Troy/Albany at the head of tide. Our recent work showed that neglecting freshwater flows leads to underestimation of peak water levels at up-river sites and neglecting stratification (typical with two-dimensional modeling) leads to underestimation all along the Hudson. As a result, we use a three-dimensional hydrodynamic model and merge streamflows and storm tides from tropical and extratropical cyclones (TCs, ETCs), as well as wet extratropical cyclone (WETC) floods (e.g. freshets, rain-on-snow events). We validate the modeled flood levels and quantify error with comparisons to 76 historical events. A Bayesian statistical method is developed for tropical cyclone streamflows using historical data and consisting in the evaluation of (1) the peak discharge and its pdf as a function of TC characteristics, and (2) the temporal trend of the hydrograph as a function of temporal evolution of the cyclone track, its intensity and the response characteristics of the specific basin. A k-nearest-neighbors method is employed to determine the hydrograph shape. Out of sample validation tests demonstrate the effectiveness of the method. Thus, the combined effects of storm surge and runoff produced by tropical cyclones hitting the New York area can be included in flood hazard assessment. Results for the upper Hudson (Albany) suggest a dominance of WETCs, for the lower Hudson (at New York Harbor) a case where ETCs are dominant for shorter return periods and TCs are more important for longer return periods (over 150 years), and for the middle-Hudson (Poughkeepsie) a mix of all three flood events types is important. However, a possible low-bias for TC flood levels is inferred from a lower importance in the assessment results, versus historical event top-20 lists, and this will be further evaluated as these preliminary methods and results are finalized. Future funded work will quantify the influences of sea level rise and flood adaptation plans (e.g. surge barriers). It would also be valuable to examine how streamflows from tropical cyclones and wet cool-season storms will change, as this factor will dominate at upriver locations.
The reduced basis method is used to construct a “universal” basis of Dirac orbitals that may be applicable throughout the nuclear chart to calibrate covariant energy density functionals. Relative to the successful development of a reduced basis emulator for the nonrelativistic Schrödinger equation, the Dirac equation adds an extra layer of complexity due to the existence of negative energy states, which complicates building an efficient reduced basis. However, once this problem is mitigated, the resulting reduced basis is able to accurately and efficiently reproduce the high-fidelity model at a fraction of the computational cost. We are confident that the resulting reduced basis will serve as a foundational element in developing rapid and accurate emulators. In turn, these emulators will play a critical role in the Bayesian optimization of covariant energy density functionals.
Estimation of isotherm model parameters from experimental data is necessary in adsorption process modeling. To facilitate isotherm modeling, databases that store experimental data exist. Nevertheless, data inconsistency among different researchers must be resolved to find a single set of isotherm parameters from multiple data sets. Herein, we propose a hierarchical Bayesian estimation method to quantify the discrepancy by a multiplicative factor while simultaneously obtaining the probability distributions of a single set of isotherm parameters. This approach also allows us to identify outliers, which are data sets that are not in agreement with other sets. The proposed approach is demonstrated for the case of CO 2 adsorption in the metal-organic framework UiO-66.
Atmospheric scientists use different methods for interpreting satellite data. In the early days of satellite meteorology, the analysis of cloud pictures from satellites was primarily subjective. As computer technology improved, satellite pictures could be processed digitally, and mathematical algorithms were developed and applied to the digital images in different wavelength bands to extract information about the atmosphere in an objective way. The kind of mathematical algorithm one applies to satellite data may depend on the complexity of the physical processes that lead to the observed image, and how much information is contained in the satellite images both spatially and at different wavelengths. Imagery from satellite-borne passive microwave radiometers has limited horizontal resolution, and the observed microwave radiances are the result of complex physical processes that are not easily modeled. For this reason, a type of algorithm called a Bayesian estimation method is utilized to interpret passive microwave imagery in an objective, yet computationally efficient manner.