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.
CASM Monte Carlo: Calculations of the thermodynamic and kinetic properties of complex multicomponent crystals
Monte Carlo techniques play a central role in statistical mechanics approaches that connect macroscopic thermodynamic and kinetic properties to the electronic structure of a material. This paper describes the implementation of Monte Carlo techniques for the study of multicomponent crystalline materials within the Clusters Approach to Statistical Mechanics (CASM) software suite, and demonstrates their use in model systems to calculate free energies and kinetic coefficients, study phase transitions, and construct phase diagrams from first principles. Many crystal structures are complex, with multiple sublattices occupied by differing sets of chemical species, along with the presence of vacancies or interstitial species. This imposes constraints on concentration variables, the form of thermodynamic potentials, and the values of kinetic transport coefficients. The framework used by CASM to formulate thermodynamic potentials and kinetic transport coefficients accounting for arbitrarily complex crystal structures is presented and demonstrated with examples of increasing complexity. Additionally, an overview of the capabilities of the CASM software specific to Monte Carlo methods is given, and a new CASM software package is introduced, casm-flow, which helps automate the setup, submission, management, and analysis of Monte Carlo simulations.
SU-F-T-349: Dosimetric Comparison of Three Different Simultaneous Integrated Boost Irradiation Techniques for Multiple Brain Metastases: Intensity-Modulatedradiotherapy, Hybrid Intensity-Modulated Radiotherapy and Volumetric Modulated Arc Therapy
Purpose: To study the dosimetric impact of intensity-modulated radiotherapy (IMRT), hybrid intensity-modulated radiotherapy (h-IMRT) and volumetric modulated arc therapy(VMAT) for whole-brain radiotherapy (WBRT) with simultaneous integrated boost in patients with multiple brain metastases. Methods: Ten patients with multiple brain metastases were included in this analysis. The prescribed dose was 45 Gy to the whole brain (PTVWBRT) and 55 Gy to individual brain metastases (PTVboost) delivered simultaneously in 25 fractions. Three treatment techniques were designed: the 7 equal spaced fields IMRT plan, hybrid IMRT plan and VMAT with two 358°arcs. In hybrid IMRT plan, two fields(90°and 270°) were planned to the whole brain. This was used as a base dose plan. Then 5 fields IMRT plan was optimized based on the two fields plan. The dose distribution in the target, the dose to the organs at risk and total MU in three techniques were compared. Results: For the target dose, conformity and homogeneity in PTV, no statistically differences were observed in the three techniques. For the maximum dose in bilateral lens and the mean dose in bilateral eyes, IMRT and h-IMRT plans showed the highest and lowest value respectively. No statistically significant differences were observed in the dose of optic nerve and brainstem. For the monitor units, IMRT and VMAT plans showed the highest and lowest value respectively. Conclusion: For WBRT with simultaneous integrated boost in patients with multiple brain metastases, hybrid IMRT could reduce the doses to lens and eyes. It is feasible for patients with brain metastases.
Statistical coverage for supersymmetric parameter estimation: a case study with direct detection of dark matter
Models of weak-scale supersymmetry offer viable dark matter (DM) candidates. Their parameter spaces are however rather large and complex, such that pinning down the actual parameter values from experimental data can depend strongly on the employed statistical framework and scanning algorithm. In frequentist parameter estimation, a central requirement for properly constructed confidence intervals is that they cover true parameter values, preferably at exactly the stated confidence level when experiments are repeated infinitely many times. Since most widely-used scanning techniques are optimised for Bayesian statistics, one needs to assess their abilities in providing correct confidence intervals in terms of the statistical coverage. Here we investigate this for the Constrained Minimal Supersymmetric Standard Model (CMSSM) when only constrained by data from direct searches for dark matter. We construct confidence intervals from one-dimensional profile likelihoods and study the coverage by generating several pseudo-experiments for a few benchmark sets of pseudo-true parameters. We use nested sampling to scan the parameter space and evaluate the coverage for the benchmarks when either flat or logarithmic priors are imposed on gaugino and scalar mass parameters. The sampling algorithm has been used in the configuration usually adopted for exploration of the Bayesian posterior. We observe both under- and over-coverage, which in some cases vary quite dramatically when benchmarks or priors are modified. We show how most of the variation can be explained as the impact of explicit priors as well as sampling effects, where the latter are indirectly imposed by physicality conditions. For comparison, we also evaluate the coverage for Bayesian credible intervals, and observe significant under-coverage in those cases.
Departures from local thermodynamic equilibrium in cutting arc plasmas derived from electron and gas density measurements using a two-wavelength quantitative Schlieren technique
A two-wavelength quantitative Schlieren technique that allows inferring the electron and gas densities of axisymmetric arc plasmas without imposing any assumption regarding statistical equilibrium models is reported. This technique was applied to the study of local thermodynamic equilibrium (LTE) departures within the core of a 30 A high-energy density cutting arc. In order to derive the electron and heavy particle temperatures from the inferred density profiles, a generalized two-temperature Saha equation together with the plasma equation of state and the quasineutrality condition were employed. Factors such as arc fluctuations that influence the accuracy of the measurements and the validity of the assumptions used to derive the plasma species temperature were considered. Significant deviations from chemical equilibrium as well as kinetic equilibrium were found at elevated electron temperatures and gas densities toward the arc core edge. An electron temperature profile nearly constant through the arc core with a value of about 14000-15000 K, well decoupled from the heavy particle temperature of about 1500 K at the arc core edge, was inferred.
Design and performance of a scalable, parallel statistics toolkit.
Most statistical software packages implement a broad range of techniques but do so in an ad hoc fashion, leaving users who do not have a broad knowledge of statistics at a disadvantage since they may not understand all the implications of a given analysis or how to test the validity of results. These packages are also largely serial in nature, or target multicore architectures instead of distributed-memory systems, or provide only a small number of statistics in parallel. This paper surveys a collection of parallel implementations of statistics algorithm developed as part of a common framework over the last 3 years. The framework strategically groups modeling techniques with associated verification and validation techniques to make the underlying assumptions of the statistics more clear. Furthermore it employs a design pattern specifically targeted for distributed-memory parallelism, where architectural advances in large-scale high-performance computing have been focused. Moment-based statistics (which include descriptive, correlative, and multicorrelative statistics, principal component analysis (PCA), and k-means statistics) scale nearly linearly with the data set size and number of processes. Entropy-based statistics (which include order and contingency statistics) do not scale well when the data in question is continuous or quasi-diffuse but do scale well when the data is discrete and compact. We confirm and extend our earlier results by now establishing near-optimal scalability with up to 10,000 processes.
The self-assembly of particles with isotropic interactions: Using DNA coated colloids to create designer nanomaterials
Self-consistent field theory equations are presented that are suitable for use as a coarse-grained model for DNA coated colloids, polymer-grafted nanoparticles and other systems with approximately isotropic interactions. The equations are generalized for arbitrary numbers of chemically distinct colloids. The advantages and limitations of such a coarse-grained approach for DNA coated colloids are discussed, as are similarities with block copolymer self-assembly. In particular, preliminary results for three species self-assembly are presented that parallel results from a two dimensional ABC triblock copolymer phase. The possibility of incorporating crystallization, dynamics, inverse statistical mechanics and multiscale modelling techniques are discussed.
Studying Velocity Turbulence from Doppler-broadened Absorption Lines: Statistics of Optical Depth Fluctuations
We continue our work on developing techniques for studying turbulence with spectroscopic data. We show that Doppler-broadened absorption spectral lines, in particular, saturated absorption lines, can be used within the framework of the previously introduced technique termed the velocity coordinate spectrum (VCS). The VCS relates the statistics of fluctuations along the velocity coordinate to the statistics of turbulence; thus, it does not require spatial coverage by sampling directions in the plane of the sky. We consider lines with different degree of absorption and show that for lines of optical depth less than one, our earlier treatment of the VCS developed for spectral emission lines is applicable, if the optical depth is used instead of intensity. This amounts to correlating the logarithms of absorbed intensities. For larger optical depths and saturated absorption lines, we show that only wings of the line are available for the analysis. In terms of the VCS formalism, this results in introducing an additional window, whose size decreases with the increase of the optical depth. As a result, strongly saturated absorption lines only carry the information about the small-scale turbulence. Nevertheless, the contrast of the fluctuations corresponding to the small-scale turbulence increases with the increase of the optical depth, which provides advantages for studying turbulence by combining lines with different optical depths. By combining different absorption lines one can develop a tomography of the turbulence in the interstellar gas in all its complexity.
A Monte Carlo technique to model performance of streak camera-based time-resolving x-ray spectrometers
A Monte Carlo technique has been developed to simulate the expected signal and the statistical noise of x-ray spectrometers that use streak cameras to achieve the time resolution required for ultrafast diagnostics of laser-generated plasmas. The technique accounts for statistics from both the photons incident on the streak camera’s photocathode and the electrons emitted by the photocathode travelling through the camera’s electron optics to the sensor. We use the technique to optimize the design of a spectrometer, which deduces the temporal history of electron temperature of the hotspot in an inertial confinement fusion implosion from its hard x-ray continuum emission spectra. The technique is general enough to be applied to any instrument using an x-ray streak camera.
Radiofrequency Ablation of Colorectal Liver Metastases: Small Size Favorably Predicts Technique Effectiveness and Survival
The objective of this study was to analyze long-term results of radiofrequency thermal ablation (RFA) for colorectal metastases (MTS), in order to evaluate predictors for adverse events, technique effectiveness, and survival. One hundred ninety-nine nonresectable MTS (0.5-8 cm; mean, 2.9 cm) in 122 patients underwent a total of 166 RFA sessions, percutaneously or during surgery. The technique was 'simple' or 'combined' with vascular occlusion. The mean follow-up time was 24.2 months. Complications, technique effectiveness, and survival rates were statistically analyzed. Adverse events occurred in 8.1% of lesions (major complication rate: 1.1%), 7.1% with simple and 16.7% with combined technique (p = 0.15). Early complete response was obtained in 151 lesions (81.2%), but 49 lesions (26.3%) recurred locally after a mean of 10.4 months. Sustained complete ablation was achieved in 66.7% of lesions {<=}3 cm versus 33.3% of lesions >3 cm (p < 0.0001). Survival rates at 1, 3, and 5 years were 91%, 54%, and 33%, respectively, from the diagnosis of MTS and 79%, 38%, and 22%, respectively, from RFA. Mean survival time from RFA was 31.5 months, 36.2 in patients with main MTS {<=}3 cm and 23.2 in those with at least one lesion >3 cm (p = 0.006). We conclude that 'simple' RFA is safe and successful for MTS {<=}3 cm, contributing to prolong survival when patients can be completely treated.
A Comparison of Techniques for Detecting Abnormal Change in Blogs
Distributed collections are made of metadata entries that contain references to artifacts not controlled by the collection curators. These collections often have limited forms of change; for digital distributed collections, primarily creation and deletion of additional resources. However, there exists a class of digital collection that undergoes additional kinds of change. These collections consist of resources that are distributed across the Internet and brought together via hyperlinking. Resources in these collections can be expected to change as time goes on. Part of the difficulty in maintaining these collections is determining if a changed page is still a valid member of the collection. Others have tried to address this by defining a maximum allowed threshold of change, however, these methods treat change as a potential problem and treat web content as static despite its intrinsic dynamicism. Instead we acknowledge change on the web as a normal part of a web document and determine the difference between what a maintainer expects a page to do and what it actually does. In this work we evaluate options for extractors and analyzers from a suite of techniques against a human-generated ground-truth set of blog changes. The results of this work show a statistically significant improvement over traditional threshold techniques for our collection.
SU-F-T-437: 3 Field VMAT Technique for Irradiation of Large Pelvic Tumors
Purpose: VMAT treatment planning for large pelvic volume irradiation could be suboptimal due to inability of Varian linac to split MLC carriage during VMAT delivery for fields larger than 14.5cm in X direction (direction of leaf motion). We compare the dosimetry between 3 VMAT planning techniques, two 2-arc field techniques and a 3-arc field technique: a) two small in X direction (less than 14.5cm) arc fields, complementing each other to cover the whole lateral extent of target during gantry rotation, b) two large arc fields, each covering the targets completely during the rotation, c) a 3 field technique with 2 small in X direction arcs and 1 large field covering whole target. Methods: 5 GYN cancer patients were selected to evaluate the 3 VMAT planning techniques. Treatment plans were generated using Varian Eclipse (ver. 11) TPS. Dose painting technique was used to deliver 5300 cGy to primary target and 4500 cGy to pelvic/abdominal node target. All the plans were normalized so that the prescription dose of 5300 cGy covered 95% of primary target volume. PTV and critical structures DVH curves were compared to evaluate all 3 planning techniques. Results: The dosimetric differences between the two 2-arc techniques were minor. The small field 2-arc technique showed a colder hot spot (0.4% averaged), while variations in maximum doses to critical structures were statistically nonsignificant (under 1.3%). In comparison, the 3-field technique demonstrated a colder hot spot (1.1% less, 105.8% averaged), and better sparing of critical structures. The maximum doses to larger bowel, small bowel and gluteal fold were 3% less, cord/cauda sparing was 4.2% better, and bladder maximum dose was 4.6% less. The differences in maximum doses to stomach and rectum were statistically nonsignificant. Conclusion: 3-arc VMAT technique for large field irradiation of pelvis demonstrates dosimetric advantages compared to 2-arc VMAT techniques.
Blade Load Unsteadiness and Turbulence Statistics in an Actuator-Line Computed Turbine–Turbine Interaction Problem
The objective of this study is to investigate how different volumetric projection techniques used in actuator-line modeling affect the unsteady blade loads and wake turbulence statistics. The two techniques for the body-force projection radius are based on either (i) the grid spacing or (ii) the combination of grid spacing and an equivalent elliptic blade planform. An array of two National Renewable Energy Laboratory 5-MW turbines separated by seven rotor diameters is simulated for 2000s (about rotor 300 revolutions) within a large-eddy simulation (LES) solver of the neutral and moderately convective atmospheric boundary layer (ABL). The statistics of sectional angle of attack (AOA), blade loads, and turbine power histories are quantified. Moreover, the degree of unsteadiness of sectional blade loads in response to atmospheric and wake turbulence is computed via a reduced frequency based on the rate-of-change in sectional AOA. Finally, the goal of this work is to make the wind energy community aware of the uncertainties associated with actuator-line modeling approaches.
Synchrophasor Data Correction under GPS Spoofing Attack: A State Estimation-Based Approach
GPS spoofing attack (GSA) has been shown to be one of the most imminent threats to almost all cyber-physical systems incorporated with the civilian GPS signal. Specifically, for our current agenda of the modernization of the power grid, this may greatly jeopardize the benefits provided by the pervasively installed phasor measurement units (PMU). In this paper, we consider the case where synchrophasor data from PMUs are compromised due to the presence of a single GSA, and show that it can be corrected by signal processing techniques. In particular, we introduce a statistical model for synchrophasorbased power system state estimation (SE), and then derive the spoofing-matched algorithms for synchrophasor data correction against GPS spoofing attack. Different testing scenarios in IEEE 14-, 30-, 57-, 118-bus systems are simulated to show the proposed algorithms’ performance on GSA detection and state estimation. Numerical results demonstrate that our proposed algorithms can consistently locate and correct the spoofed synchrophasor data with good accuracy as long as the system observability is satisfied. The accuracy of state estimation is significantly improved compared with the traditional weighted least square method and approaches the performance under the Genie-aided method.
A Hybrid dasymetric and machine learning approach to high-resolution residential electricity consumption modeling
As urban areas continue to grow and evolve in a world of increasingenvironmental awareness, the need for detailed information regarding residentialenergy consumption patterns has become increasingly important. Though currentmodeling efforts mark significant progress in the effort to better understand thespatial distribution of energy consumption, the majority of techniques are highlydependent on region-specific data sources and often require building- ordwelling-level details that are not publicly available for many regions in the UnitedStates. Furthermore, many existing methods do not account for errors in input datasources and may not accurately reflect inherent uncertainties in model outputs. Wepropose an alternative and more general hybrid approach to high-resolution residentialelectricity consumption modeling by merging a dasymetric model with acomplementary machine learning algorithm. The method’s flexible data requirementand statistical framework ensure that the model both is applicable to a widerange of regions and considers errors in input data sources.
THE OBSERVATIONS OF REDSHIFT EVOLUTION IN LARGE-SCALE ENVIRONMENTS (ORELSE) SURVEY. I. THE SURVEY DESIGN AND FIRST RESULTS ON CL 0023+0423 AT z = 0.84 AND RX J1821.6+6827 AT z = 0.82
We present the Observations of Redshift Evolution in Large-Scale Environments (ORELSE) Survey, a systematic search for structure on scales greater than 10 h {sup -1} {sub 70} Mpc around 20 well-known clusters at redshifts of 0.6 < z < 1.3. The goal of the survey is to examine a statistical sample of dynamically active clusters and large-scale structures in order to quantify galaxy properties over the full range of local and global environments. We describe the survey design, the cluster sample, and our extensive observational data covering at least 25' around each target cluster. We use adaptively smoothed red galaxy density maps from our wide-field optical imaging to identify candidate groups/clusters and intermediate-density large-scale filaments/walls in each cluster field. Because photometric techniques (such as photometric redshifts, statistical overdensities, and richness estimates) can be highly uncertain, the crucial component of this survey is the unprecedented amount of spectroscopic coverage. We are using the wide-field, multiobject spectroscopic capabilities of the Deep Multiobject Imaging Spectrograph to obtain 100-200+ confirmed cluster members in each field. Our survey has already discovered the Cl 1604 supercluster at z {approx} 0.9, a structure which contains at least eight groups and clusters and spans 13 Mpc x 100 Mpc. Here, we present the results on the large-scale environments of two additional clusters, Cl 0023+0423 at z = 0.84 and RX J1821.6+6827 at z = 0.82, which highlight the diversity of global properties at these redshifts. The optically selected Cl 0023+0423 is a four-way group-group merger with constituent groups having measured velocity dispersions between 206 and 479 km s{sup -1}. The galaxy population is dominated by blue, star-forming galaxies, with 80% of the confirmed members showing [O II] emission. The strength of the H{delta} line in a composite spectrum of 138 members indicates a substantial contribution from recent starbursts to the overall galaxy population. In contrast, the X-ray-selected RX J1821.6+6827 is a largely isolated, massive cluster with a measured velocity dispersion of 926 {+-} 77 km s{sup -1}. The cluster exhibits a well-defined red sequence with a large quiescent galaxy population. The results from these two targets, along with preliminary findings on other ORELSE clusters, suggest that optical selection may be more effective than X-ray surveys at detecting less-evolved, dynamically active systems at these redshifts.
ASTROPHYSICAL FLUID DYNAMICS VIA DIRECT STATISTICAL SIMULATION
In this paper, we introduce the concept of direct statistical simulation for astrophysical flows. This technique may be appropriate for problems in astrophysical fluids where the instantaneous dynamics of the flows are of secondary importance to their statistical properties. We give examples of such problems including mixing and transport in planets, stars, and disks. The method is described for a general set of evolution equations, before we consider the specific case of a spectral method optimized for problems on a spherical surface. The method is illustrated for the simplest non-trivial example of hydrodynamics and magnetohydrodynamics on a rotating spherical surface. We then discuss possible extensions of the method both in terms of computational methods and the range of astrophysical problems that are of interest.
The impact of anisotropic redshift distributions on angular clustering
A leading way to constrain physical theories from cosmological observations is to test their predictions for the angular clustering statistics of matter tracers, a technique that is set to become ever more central with the next generation of large imaging surveys. Interpretation of this clustering requires knowledge of the projection kernel, or the redshift distribution of the sources, and the typical assumption is an isotropic redshift distribution for the objects. However, variations in the kernel are expected across the survey footprint due to photometric variations and residual observational systematic effects. Here, we develop the formalism for anisotropic projection and present several limiting cases that elucidate the key aspects. We quantify the impact of anisotropies in the redshift distribution on a general class of angular two-point statistics. In particular, we identify a mode-coupling effect that can add power to auto-correlations, including galaxy clustering and cosmic shear, and remove it from certain cross-correlations. If the projection anisotropy is primarily at large scales, the mode-coupling depends upon its variance as a function of redshift; furthermore, it is often of similar shape to the signal. In contrast, the cross-correlation of a field whose selection function is anisotropic with another one featuring no such variations — such as CMB lensing — is immune to these effects. We discuss explicitly several special cases of the general formalism including galaxy clustering, galaxy-galaxy lensing, cosmic shear and cross-correlations with CMB lensing, and publicly release a code to compute the biases.