Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Statistical techniques”

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 127 records · Page 7

HIGH-ENERGY NEUTRINO AND GAMMA-RAY TRANSIENTS FROM TRANS-RELATIVISTIC SUPERNOVA SHOCK BREAKOUTS

Trans-relativistic shocks that accompany some supernovae (SNe) produce X-ray burst emissions as they break out in the dense circumstellar medium around the progenitors. This phenomenon is sometimes associated with peculiar low-luminosity gamma-ray bursts (LL GRBs). Here, we investigate the high-energy neutrino and gamma-ray counterparts of such a class of SNe. Just beyond the shock breakout radius, particle acceleration in the collisionless shock starts to operate in the presence of breakout photons. We show that protons may be accelerated to sufficiently high energies and produce high-energy neutrinos and gamma rays via the photomeson interaction. These neutrinos and gamma rays may be detectable from {approx}< 10 Mpc away by IceCube/KM3Net as multi-TeV transients almost simultaneously with the X-ray breakout, and even from {approx}< 100 Mpc away with follow-up observations by the Cherenkov Telescope Array using a wide-field sky monitor like Swift as a trigger. A statistical technique using a stacking approach could also be possible for the detection, with the aid of the SN optical/infrared counterparts. Such multi-messenger observations offer the possibility to probe the transition of trans-relativistic shocks from radiation-mediated to collisionless ones, and would also constrain the mechanisms of particle acceleration and emission in LL GRBs.

79 ASTRONOMY AND ASTROPHYSICS↗

A best-practice guide to predicting plant traits from leaf-level hyperspectral data using partial least squares regression

Partial least squares regression (PLSR) modelling is a statistical technique for correlating datasets, and involves the fitting of a linear regression between two matrices. One application of PLSR enables leaf traits to be estimated from hyperspectral optical reflectance data, facilitating rapid, high-throughput, non-destructive plant phenotyping. This technique is of interest and importance in a wide range of contexts including crop breeding and ecosystem monitoring. The lack of a consensus in the literature on how to perform PLSR means that interpreting model results can be challenging, applying existing models to novel datasets can be impossible, and unknown or undisclosed assumptions can lead to incorrect or spurious predictions. We address this lack of consensus by proposing best practices for using PLSR to predict plant traits from leaf-level hyperspectral data, including a discussion of when PLSR is applicable, and recommendations for data collection. Further, we provide a tutorial to demonstrate how to develop a PLSR model, in the form of an R script accompanying this manuscript. This practical guide will assist all those interpreting and using PLSR models to predict leaf traits from spectral data, and advocates for a unified approach to using PLSR for predicting traits from spectra in the plant sciences.

54 ENVIRONMENTAL SCIENCES↗

Interpolating between small- and large-$\mathscr{g}$ expansions using Bayesian model mixing

Bayesian model mixing (BMM) is a statistical technique that can be used to combine models that are predictive in different input domains into a composite distribution that has improved predictive power over the entire input space. We explore the application of BMM to the mixing of two expansions of a function of a coupling constant $\mathscr{g}$ that are valid at small and large values of $\mathscr{g}$ respectively. This type of problem is quite common in nuclear physics, where physical properties are straightforwardly calculable in strong and weak interaction limits or at low and high densities or momentum transfers, but difficult to calculate in between. Interpolation between these limits is often accomplished by a suitable interpolating function, e.g., Padé approximants, but it is then unclear how to quantify the uncertainty of the interpolant. We address this problem in the simple context of the partition function of zero-dimensional Φ 4 theory, for which the (asymptotic) expansion at small $\mathscr{g}$ and the (convergent) expansion at large $\mathscr{g}$ are both known. We consider three mixing methods: linear mixture BMM, localized bivariate BMM, and localized multivariate BMM with Gaussian processes. Here, we find that employing a Gaussian process in the intermediate region between the two predictive models leads to the best results of the three methods. The methods and validation strategies we present here should be generalizable to other nuclear physics settings.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

α -decay energies of superheavy nuclei: Systematic trends

Background: New superheavy nuclei are often identified through their characteristic α -decay energies, which requires accurate calculations of Q α values. While many Q α predictions are available, little is known about their uncertainties, and this makes it difficult to carry out extrapolations to as-yet-unknown systems. Purpose: This work aims to analyze several models, compare their predictions to available experimental data, and study their performance for the unobserved α-decay chains of 296 120 and 298 120, which are of current experimental interest. Our quantified results will also serve as a benchmark for future, more sophisticated statistical studies. Methods: We use nuclear superfluid density functional theory (DFT) with several Skyrme energy density functionals (EDFs). To estimate systematic model uncertainties, we employ uniform model averaging. Results: We investigated the Q α values for even-even nuclei from Fm to Z = 120 . For well-deformed nuclei between Fm and Ds, we find excellent consistency between different model predictions, and a good agreement with experimental results. For transitional nuclei beyond Ds, intermodel differences grow, resulting in an appreciable systematic error. In particular, our models underestimate Q α for the heaviest nucleus 294 Og. Conclusions The robustness of DFT predictions for well-deformed superheavy nuclei supports the idea of using experimental Q α values, together with theoretical predictions, as reasonable ( Z, A ) indicators. Unfortunately, this identification method is not projected to work well in the region of deformed-to-spherical shape transition as one approaches N = 184 . The use of Q α values in the identification of new superheavy nuclei will benefit greatly from progress in developing both new spectroscopic-quality EDFs and more sophisticated statistical techniques of uncertainty quantification.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Lattice quantum chromodynamics at large isospin density

We present an algorithm to compute correlation functions for systems with the quantum numbers of many identical mesons from lattice quantum chromodynamics (QCD). The algorithm is numerically stable and allows for the computation of n-pion correlation functions for n ϵ {1, … , N} using a single N × N matrix decomposition, improving on previous algorithms. We apply the algorithm to calculations of correlation functions with up to 6144 charged pions using two ensembles of gauge field configurations generated with quark masses corresponding to a pion mass m π = 170 MeV and spacetime volumes of (4.4 3 × 8.8) fm 4 and (5.8 3 × 11.6) fm 4 . We also discuss statistical techniques for the analysis of such systems, in which the correlation functions vary over many orders of magnitude. In particular, we observe that the many-pion correlation functions are well-approximated by log-normal distributions, allowing the extraction of the energies of these systems. Using these energies, the large-isospin-density, zero-baryon-density region of the QCD phase diagram is explored. A peak is observed in the energy density at an isospin chemical potential μ I ~ 1.5m π , signaling the transition into a Bose-Einstein condensed phase. The isentropic speed of sound, c s , in the medium is seen to exceed the ideal-gas (conformal) limit ($c^{2}_{s} ≤ 1/3)$ over a wide range of chemical potential before falling towards the asymptotic expectation at μ I ~ 15m π . These, and other thermodynamic observables, indicate that the isospin chemical potential must be large for the system to be well described by an ideal gas or perturbative QCD.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Dark matter constraints from a joint analysis of dwarf Spheroidal galaxy observations with VERITAS

We present constraints on the annihilation cross section of weakly interacting massive particles dark matter based on the joint statistical analysis of four dwarf galaxies with VERITAS. These results are derived from an optimized photon weighting statistical technique that improves on standard imaging atmospheric Cherenkov telescope (IACT) analyses by utilizing the spectral and spatial properties of individual photon events.

79 ASTRONOMY AND ASTROPHYSICS↗

Constraining the equation of state of superhadronic matter from heavy-ion collisions

The equation of state of QCD matter for temperatures near and above the quark-hadron transition (~165 MeV) is inferred within a Bayesian framework through the comparison of data from the Relativistic Heavy Ion Collider and from the Large Hadron Collider to theoretical models. State-of-the-art statistical techniques are applied to simultaneously analyze multiple classes of observables while varying 14 independent model parameters. Thus, the resulting posterior distribution over possible equations of state is consistent with results from lattice gauge theory.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

A Framework for Counterfeit Smart Grid Device Detection

The core vision of the smart grid concept is the realization of reliable two-­way communications between smart devices (e.g., IEDs, PLCs, PMUs). The benefits of the smart grid also come with tremendous security risks and new challenges in protecting the smart grid systems from cyber threats. Particularly, the use of untrusted counterfeit smart grid devices represents a real problem. Consequences of propagating false or malicious data, as well as stealing valuable user or smart grid state information from counterfeit devices are costly. Hence, early detection of counterfeit devices is critical for protecting smart grid’s components and users. To address these concerns, in this poster, we introduce our initial design of a configurable framework that utilize system call tracing, library interposition, and statistical techniques for monitoring and detection of counterfeit smart grid devices. In our framework, we consider six different counterfeit device scenarios with different smart grid devices and adversarial seZings. Our initial results on a realistic testbed utilizing actual smart-­grid GOOSE messages with IEC-­61850 communication protocol are very promising. Our framework is showing excellent rates on detection of smart grid counterfeit devices from impostors.

97 MATHEMATICS AND COMPUTING↗

Nonparametric Multivariate Anomaly Analysis in Support of HPC Resilience

Large-scale computing systems provide great potential for scientific exploration. However, the complexity that accompanies these enormous machines raises challeges for both, users and operators. The effective use of such systems is often hampered by failures encountered when running applications on systems containing tens-of-thousands of nodes and hundreds-of-thousands of compute cores capable of yielding petaflops of performance. In systems of this size failure detection is complicated and root-cause diagnosis difficult. This paper describes our recent work in the identification of anomalies in monitoring data and system logs to provide further insights into machine status, runtime behavior, failure modes and failure root causes. It discusses the details of an initial prototype that gathers the data and uses statistical techniques for analysis.

97 MATHEMATICS AND COMPUTING↗

Time-of-flight neutron spectrum unfolding for mixed-oxide nuclear fuel and plutonium metal using a dual-particle imager

A dual-particle imager with sensitivity to both neutrons and photons has been developed for the detection of special nuclear material. This system has the capability to provide spectral information on detected sources that is useful for safeguard applications. In nuclear facilities with high photon background rates, neutron detection can aid in identification and verification capabilities. The dual-particle imager creates a reconstructed neutron spectrum by correlating neutron counts in pairs of liquid scintillators. The reconstructed spectrum, a combination of energy deposited and time-of-flight, contains resolution effects that blur the true, emitted source-spectrum. To reduce the impact of resolution effects, the reconstructed spectrum can be unfolded to create an estimated source spectrum. A statistical technique, maximum-likelihood expectation-maximization, has been shown to produce an unfolded result for Cf-252 and Am-Be. This technique requires a system response-matrix that contains reconstructed spectra for single neutron energies. Since measuring mono-energetic neutrons is difficult, the system-response-matrix is simulated. Monte Carlo modeling with MCNPX-PoliMi has been show to produce an accurate system response. This work expands on previous neutron unfolding results by showing results for mixed-oxide fuel and plutonium metal measured with the dual-particle imager. Furthermore these results also demonstrate the ability of the dual-particle imager to discrimination between spontaneous fission neutrons and (α,n) reaction neutrons.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Characterization and differentiation of aluminum powders used in improvised explosive devices. Part 2: Micromorphometric method refinement and preliminary statistical analysis

Abstract Aluminum (Al) powder is commonly encountered in improvised explosive devices (IEDs) as a metallic fuel due to its availability and low cost. Although available commercially in powder form, amateur bomb‐makers also produce their own Al powder via simple methods found online. In order to provide investigative leads and forensic intelligence, it is important to evaluate not only the composition of homemade devices, but also to distinguish between the various forms of Al powder they contain. To achieve this goal, a method using automated microscopy in combination with statistical techniques has been demonstrated to have the potential to provide source discrimination and investigative leads in source attribution of Al powders in IEDs. The present research refined this method and investigated 59 industrially and amateurly produced Al powder sources with seven subsamples per source using two traditional linear discriminant analyses (LDA), one with a standard data split for training and testing, and another using leave‐one‐out cross‐validation. Averaging the classification accuracies for the two LDA‐based analyses, LDA has the ability to correctly classify 59.26%, 83.35%, and 80.69% of the samples based on their powder source, type, and production method, respectively. This classification accuracy represents a 3407%, 317%, and 61.38% increase in accuracy from random class assignment, respectively. Further, in most instances of incorrect data attribution to a particular source, the subsample has been misidentified with another sample of the same powder type or production method.

Ommen, Danica M.↗

Effects of sample design and landscape features on a measure of environmental heterogeneity

Environmental heterogeneity, an important influence on organisms and ecological processes, can be quantified by the variance of an environmental characteristic over all locations within a study extent. However on landscapes with autocorrelation and gradient patterns, estimating this variance from a sample of locations may lead to errors that cannot be corrected with statistical techniques. We analytically derived the relative expected sampling error of sample designs on landscapes with particular gradient pattern and autocorrelation features. We applied this closed-form approach to temperature observations from an existing study. The expected heterogeneity differed, both in magnitude and direction, amongst sample designs over the study site's likely range of autocorrelation and gradient features. We conducted a simulation study to understand the effects of (i) landscape variability and (ii) design variability on an average sampling error. On 10 000 simulated landscapes with varying gradient and autocorrelation features, we compared estimates of variance from a variety of structured and random sample designs. While gradient patterns and autocorrelation cause large errors for some designs, others yield near-zero average sampling error. Sample location spacing is a key factor in sample design performance. Random designs have larger range of possible sampling errors than structured designs due to the potential for sample arrangements that over- and under-sample certain areas of the landscape. Furthermore, when implementing a new sample design to quantify environmental heterogeneity via variance, we recommend using a simple structured design with appropriate sample spacing. For existing designs, we recommend calculating the relative expected sampling error via our analytical derivation.

54 ENVIRONMENTAL SCIENCES↗

Modeling and forecasting the distribution of Vibrio vulnificus in Chesapeake Bay

The aim is to construct statistical models to predict the presence, abundance and potential virulence of Vibrio vulnificus in surface waters. A variety of statistical techniques were used in concert to identify water quality parameters associated with V. vulnificus presence, abundance and virulence markers in the interest of developing strong predictive models for use in regional oceanographic modeling systems. A suite of models are provided to represent the best model fit and alternatives using environmental variables that allow them to be put to immediate use in current ecological forecasting efforts. Conclusions: Environmental parameters such as temperature, salinity and turbidity are capable of accurately predicting abundance and distribution of V. vulnificus in Chesapeake Bay. Forcing these empirical models with output from ocean modeling systems allows for spatially explicit forecasts for up to 48 h in the future. This study uses one of the largest data sets compiled to model Vibrio in an estuary, enhances our understanding of environmental correlates with abundance, distribution and presence of potentially virulent strains and offers a method to forecast these pathogens that may be replicated in other regions.

Jacobs, John M.↗

CV4Quantum

CV4Quantum is a statistical technique for reducing the sampling overhead in probabilistic error cancellation, which is an error mitigation technique used in quantum computing. CV4Quantum is based on the control variates method, which is a Monte Carlo variance reduction technique. This repository contains the code and data associated with a demonstration of CV4Quantum using simulation experiments.

Shyamsundar, Prasanth [Fermi National Accelerator ↗

Principal Component Analysis for Extremes and Application to U.S. Precipitation

Here, we propose a method for analyzing extremal behavior through the lens of a most efficient basis of vectors. The method is analogous to principal component analysis, but is based on methods from extreme value analysis. Specifically, rather than decomposing a covariance or correlation matrix, we obtain our basis vectors by performing an eigendecomposition of a matrix that describes pairwise extremal dependence. We apply the method to precipitation observations over the contiguous United States. We find that the time series of large coefficients associated with the leading eigenvector shows very strong evidence of a positive trend, and there is evidence that large coefficients of other eigenvectors have relationships with El Niño–Southern Oscillation.

54 ENVIRONMENTAL SCIENCES↗

Bioeconomy Survey Results Regarding Barriers to the United States Advanced Biofuel Industry

Although the 2005 Environmental Protection Act (EPAct) was enacted to bolster the emerging biofuel industry, 52% of advanced biofuel (AB) projects ended by 2015. However, there are no complete lists of internal and external barriers that can help to explain why these projects are failing. The goal of this study was to develop a list of barriers impeding advanced biofuel projects by conducting a survey of biofuel stakeholders. Based on a literature review and previous research, a list of 23 hypothesized internal and external barriers was elaborated. A survey was conducted to have industry stakeholders provide their perception on the list of hypothesized barriers. The perceptions of industry stakeholders were analyzed by dividing the sample in three different stakeholder groups: advanced biofuel industry members, government representatives, and a third category called others that included publishers, journalists, suppliers, and other related stakeholders to the industry. In addition, nonparametric statistical techniques were used to compare the perceptions of the groups. Here, the most significant results indicated that Technology issues was considered as an internal barrier for the three groups while Funding and Renewable Fuel Standards were perceived as external barriers by the three groups too. In addition, the rating of barriers was further analyzed only by AB industry stakeholders in order to uncover more details on the perception of barriers that might be preventing the AB industry to prosper.

09 BIOMASS FUELS↗

Hourly gap-filled meteorological data from PIE LTER measurements (2004-2023) used as drivers to run ELM PFLOTRAN simulations

This dataset contains continuous gap-filled precipitation, solar radiation, photosynthetically active radiation (PAR), air temperature, relative humidity, wind speed, and barometric pressure data recorded primarily at the Marshview Farm weather station within the Plum Island Long Term Ecosystems Research (PIE LTER) in Newbury Massachusetts (MA) from 2004 to 2023. We compiled the data set from published annual data packages in 15min resolution available on DataOne. Gaps were filled using different statistical techniques or available observations from the vicinity, e.g. the US-PLo and the US-PHM Ameriflux sites, also located within the PIE LTER. Flags are included in this dataset to indicate the origin of each data point. Metadata files ELMPFLOTRAN_met_dd.csv and ELMPFLOTRAN_met_flmd.csv contain more information on site locations, gap filling protocols, data variables, flags, and QA/QC methods. The data set was used in the spin up and simulations of a land surface model coupled to a biogeochemical reaction network (ELM PFLOTRAN) assessing impacts of hydrology and salinity input on methane fluxes in 2022 and 2023 (Sulman et al., 2024).

54 ENVIRONMENTAL SCIENCES↗

Tracking the Sun III; The Installed Cost of Photovoltaics in the United States from 1998-2009

Installations of solar photovoltaic (PV) systems have been growing at a rapid pace in recent years. In 2009, approximately 7,500 megawatts (MW) of PV were installed globally, up from approximately 6,000 MW in 2008, consisting primarily of grid-connected applications. With 335 MW of grid-connected PV capacity added in 2009, the United States was the world's fourth largest PV market in 2009, behind Germany, Italy, and Japan. The market for PV in the United States is driven by national, state, and local government incentives, including up-front cash rebates, production-based incentives, requirements that electricity suppliers purchase a certain amount of solar energy, and federal and state tax benefits. These programs are, in part, motivated by the popular appeal of solar energy, and by the positive attributes of PV - modest environmental impacts, avoidance of fuel price risks, coincidence with peak electrical demand, and the possible deployment of PV at the point of use. Given the relatively high cost of PV, however, a key goal of these policies is to encourage cost reductions over time. Therefore, as policy incentives have become more significant and as PV deployment has accelerated, so too has the desire to track the installed cost of PV systems over time, by system characteristics, by system location, and by component. Despite the significant year-on-year growth, however, the share of global and U.S. electricity supply met with PV remains small, and annual PV additions are currently modest in the context of the overall electric system. To address this need, Lawrence Berkeley National Laboratory initiated a report series focused on describing trends in the installed cost of grid-connected PV systems in the United States. The present report, the third in the series, describes installed cost trends from 1998 through 2009, and provides preliminary cost data for systems installed in 2010. The analysis is based on project-level cost data from approximately 78,000 residential and non-residential PV systems in the U.S., all of which are installed at end-use customer facilities (herein referred to as 'customer-sited' systems). The combined capacity of systems in the data sample totals 874 MW, equal to 70% of all grid-connected PV capacity installed in the United States through 2009 and representing one of the most comprehensive sources of installed PV cost data for the U.S. The report also briefly compares recent PV installed costs in the United States to those in Germany and Japan. Finally, it should be noted that the analysis presented here focuses on descriptive trends in the underlying data, serving primarily to summarize the data in tabular and graphical form; later analysis may explore some of these trends with more-sophisticated statistical techniques. The report begins with a summary of the data collection methodology and resultant dataset (Section 2). The primary findings of the analysis are presented in Section 3, which describes trends in installed costs prior to receipt of any financial incentives: over time and by system size, component, state, system ownership type (customer-owned vs. third party-owned), host customer segment (residential vs. commercial vs. public-sector vs. non-profit), application (new construction vs. retrofit), and technology type (building-integrated vs. rack-mounted, crystalline silicon vs. thin-film, and tracking vs. fixed-axis). Section 4 presents additional findings related to trends in PV incentive levels over time and among states (focusing specifically on state and utility incentive programs as well as state and federal tax credits), and trends in the net installed cost paid by system owners after receipt of such incentives. Brief conclusions are offered in the final section, and several appendices provide additional details on the analysis methodology and additional tabular summaries of the data.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗