SEARCH · Search NASA

Results for “Variance analysis”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 595 records · Page 33

All sources NASA NTRS DOE OSTI

DOE OSTI2019

Multifidelity Uncertainty Quantification with Applications in Wind Turbine Aerodynamics: Preprint

The propagation of input uncertainty through engineering models allows designers to better understand the range of possible outcomes resulting from design decisions. This could lead to greater trust between modelers and stakeholders in the wind energy industry. In this study, we apply multilevel-multifidelity Monte Carlo sampling to flow over an airfoil, assuming uncertainty in the inflow conditions, and characterize the associated computational savings compared to standard Monte Carlo approaches. The truth model is provided by an airfoil simulation with a very fine computational time step, and auxiliary lower-level models are provided by simulations with coarser time steps. Reynolds-averaged Navier Stokes and detached eddy simulations are used to obtain two different model fidelities. The primary quantity of interest for this analysis is the lift force, which is examined for a range of angles of attack. We launch an initial set of 'trial' samples to determine the optimal allocation of model evaluations, and these trial evaluations are used to inform a larger sampling effort. Using the multilevel-multifidelity approach, we achieve roughly an order of magnitude variance reduction in expected lift as compared to the standard Monte Carlo approach with an equivalent computational cost.

17 WIND ENERGY↗

DOE OSTI2015

Dependence of Achievable Plan Quality on Treatment Technique and Planning Goal Refinement: A Head-and-Neck Intensity Modulated Radiation Therapy Application

Purpose: To develop a practical workflow for retrospectively analyzing target and normal tissue dose–volume endpoints for various intensity modulated radiation therapy (IMRT) delivery techniques; to develop technique-specific planning goals to improve plan consistency and quality when feasible. Methods and Materials: A total of 165 consecutive head-and-neck patients from our patient registry were selected and retrospectively analyzed. All IMRT plans were generated using the same dose–volume guidelines for TomoTherapy (Tomo, Accuray), TrueBeam (TB, Varian) using fixed-field IMRT (TB-IMRT) or RAPIDARC (TB-RAPIDARC), or Siemens Oncor (Siemens-IMRT, Siemens). A MATLAB-based dose–volume extraction and analysis tool was developed to export dosimetric endpoints for each patient. With a fair stratification of patient cohort, the variation of achieved dosimetric endpoints was analyzed among different treatment techniques. Upon identification of statistically significant variations, technique-specific planning goals were derived from dynamically accumulated institutional data. Results: Retrospective analysis showed that although all techniques yielded comparable target coverage, the doses to the critical structures differed. The maximum cord doses were 34.1 ± 2.6, 42.7 ± 2.1, 43.3 ± 2.0, and 45.1 ± 1.6 Gy for Tomo, TB-IMRT, TB-RAPIDARC, and Siemens-IMRT plans, respectively. Analyses of variance showed significant differences for the maximum cord doses but no significant differences for other selected structures among the investigated IMRT delivery techniques. Subsequently, a refined technique-specific dose–volume guideline for maximum cord dose was derived at a confidence level of 95%. The dosimetric plans that failed the refined technique-specific planning goals were reoptimized according to the refined constraints. We observed better cord sparing with minimal variations for the target coverage and other organ at risk sparing for the Tomo cases, and higher parotid doses for C-arm linear accelerator–based IMRT and RAPIDARC plans. Conclusion: Patient registry–based processes allowed easy and systematic dosimetric assessment of treatment plan quality and consistency. Our analysis revealed the dependence of certain dosimetric endpoints on the treatment techniques. Technique-specific refinement of planning goals may lead to improvement in plan consistency and plan quality.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

DOE OSTI2016

Application of Monte Carlo Methods to Perform Uncertainty and Sensitivity Analysis on Inverse Water-Rock Reactions with NETPATH

Methods were developed to quantify uncertainty and sensitivity for NETPATH inverse water-rock reaction models and to calculate dissolved inorganic carbon, carbon-14 groundwater travel times. The NETPATH models calculate upgradient groundwater mixing fractions that produce the downgradient target water chemistry along with amounts of mineral phases that are either precipitated or dissolved. Carbon-14 groundwater travel times are calculated based on the upgradient source-water fractions, carbonate mineral phase changes, and isotopic fractionation. Custom scripts and statistical code were developed for this study to facilitate modifying input parameters, running the NETPATH simulations, extracting relevant output, postprocessing the results, and producing graphs and summaries. The scripts read userspecified values for each constituent’s coefficient of variation, distribution, sensitivity parameter, maximum dissolution or precipitation amounts, and number of Monte Carlo simulations. Monte Carlo methods for analysis of parametric uncertainty assign a distribution to each uncertain variable, sample from those distributions, and evaluate the ensemble output. The uncertainty in input affected the variability of outputs, namely source-water mixing, phase dissolution and precipitation amounts, and carbon-14 travel time. Although NETPATH may provide models that satisfy the constraints, it is up to the geochemist to determine whether the results are geochemically reasonable. Two example water-rock reaction models from previous geochemical reports were considered in this study. Sensitivity analysis was also conducted to evaluate the change in output caused by a small change in input, one constituent at a time. Results were standardized to allow for sensitivity comparisons across all inputs, which results in a representative value for each scenario. The approach yielded insight into the uncertainty in water-rock reactions and travel times. For example, there was little variation in source-water fraction between the deterministic and Monte Carlo approaches, and therefore, little variation in travel times between approaches. Sensitivity analysis proved very useful for identifying the most important input constraints (dissolved-ion concentrations), which can reveal the variables that have the most influence on source-water fractions and carbon-14 travel times. Once these variables are determined, more focused effort can be applied to determining the proper distribution for each constraint. Second, Monte Carlo results for water-rock reaction modeling showed discrete and nonunique results. The NETPATH models provide the solutions that satisfy the constraints of upgradient and downgradient water chemistry. There can exist multiple, discrete solutions for any scenario and these discrete solutions cause grouping of results. As a result, the variability in output may not easily be represented by a single distribution or a mean and variance and care should be taken in the interpretation and reporting of results.

phase dissolution and precipitation amounts↗

DOE OSTI2016

Testing and validation of multi‐lidar scanning strategies for wind energy applications

Abstract Several factors cause lidars to measure different values of turbulence than an anemometer on a tower, including volume averaging, instrument noise and the use of a scanning circle to estimate the wind field. One way to avoid the use of a scanning circle is to deploy multiple scanning lidars and point them toward the same volume in space to collect velocity measurements and extract high‐resolution turbulence information. This paper explores the use of two multi‐lidar scanning strategies, the tri‐Doppler technique and the virtual tower technique, for measuring 3‐D turbulence. In summer 2013, a vertically profiling Leosphere WindCube lidar and three Halo Photonics Streamline lidars were operated at the Southern Great Plains Atmospheric Radiation Measurement site to test these multi‐lidar scanning strategies. During the first half of the field campaign, all three scanning lidars were pointed at approximately the same point in space and a tri‐Doppler analysis was completed to calculate the three‐dimensional wind vector every second. Next, all three scanning lidars were used to build a ‘virtual tower’ above the WindCube lidar. Results indicate that the tri‐Doppler technique measures higher values of horizontal turbulence than the WindCube lidar under stable atmospheric conditions, reduces variance contamination under unstable conditions and can measure high‐resolution profiles of mean wind speed and direction. The virtual tower technique provides adequate turbulence information under stable conditions but cannot capture the full temporal variability of turbulence experienced under unstable conditions because of the time needed to readjust the scans. Copyright © 2016 John Wiley & Sons, Ltd.

Newman, Jennifer F.↗

DOE OSTI2015

Variance-based sensitivity indices for stochastic models with correlated inputs

The goal of this article is the formulation of the principles of one of the possible strategies in implementing correlation between input random variables so as to be usable for algorithm development and the evaluation of Sobol’s sensitivity analysis. With regard to the types of stochastic computational models, which are commonly found in structural mechanics, an algorithm was designed for effective use in conjunction with Monte Carlo methods. Sensitivity indices are evaluated for all possible permutations of the decorrelation procedures for input parameters. The evaluation of Sobol’s sensitivity coefficients is illustrated on an example in which a computational model was used for the analysis of the resistance of a steel bar in tension with statistically dependent input geometric characteristics.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

DOE OSTI2018

Bias Impact Analysis and Calibration of UAV-Based Mobile LiDAR System with Spinning Multi-Beam Laser Scanner

Light Detection and Ranging (LiDAR) is a technology that uses laser beams to measure ranges and generates precise 3D information about the scanned area. It is rapidly gaining popularity due to its contribution to a variety of applications such as Digital Building Model (DBM) generation, telecommunications, infrastructure monitoring, transportation corridor asset management and crash/accident scene reconstruction. To derive point clouds with high positional accuracy, estimation of mounting parameters relating the laser scanners to the onboard Global Navigation Satellite System/Inertial Navigation System (GNSS/INS) unit, i.e., the lever-arm and boresight angles, is the foremost and necessary step. This paper proposes a LiDAR system calibration strategy for a Unmanned Aerial Vehicle (UAV)-based mobile mapping system that can directly estimate the mounting parameters for spinning multi-beam laser scanners through an outdoor calibration procedure. This approach is based on the use of conjugate planar/linear features in overlapping point clouds derived from different flight lines. Designing an optimal configuration for calibration is the first and foremost step in order to ensure the most accurate estimates of mounting parameters. This is achieved by conducting a rigorous theoretical analysis of the potential impact of bias in mounting parameters of a LiDAR unit on the resultant point cloud. The dependency of the impact on the orientation of target primitives and relative flight line configuration would help in deducing the configuration that would maximize as well as decouple the impact of bias in each mounting parameter so as to ensure their accurate estimation. Finally, the proposed analysis and calibration strategy are validated by calibrating a UAV-based LiDAR system using two different datasets—one acquired with flight lines at a single flying height and the other with flight lines at two different flying heights. With this being said, the calibration performance is evaluated by analyzing correlation between the estimated system parameters, the a-posteriori variance factor of the Least Squares Adjustment (LSA) procedure and the quality of fit of the adjusted point cloud to planar/linear features before and after the calibration process.

3D point clouds↗

DOE OSTI2020

Statistical Modeling and Analysis of k -Layer Coverage of Two-Dimensional Materials in Inkjet Printing Processes

Two-dimensional layered materials/flakes, also known as crystalline atom-thick layer nanosheets, have recently been receiving great attention in electronics fabrication due to their unique and intriguing properties. The k-layer coverage area (i.e., the area covered by k number of overlapping layers) of the printed flake pattern significantly impacts on the properties of the printed electronics. In this work, we constructed a statistical model to describe the k-layer coverage of randomly distributed two-dimensional materials. A series of results are obtained to provide not only the expectation but also the variance of the coverage area. The boundary effects on the random flakes coverage are also studied. In addition, an approximated statistical testing approach is also developed in this work to detect abnormal coverage patterns. Furthermore, the case studies based on simulated data and real flakes images obtained from the inkjet printing process demonstrate the accuracy and effectiveness of the proposed model and analysis methods.

36 MATERIALS SCIENCE↗

DOE OSTI2018

Direct Breakthrough Curve Prediction From Statistics of Heterogeneous Conductivity Fields

In this paper, we present a methodology to predict the shape of solute breakthrough curves in heterogeneous aquifers at early times and/or under high degrees of heterogeneity, both cases in which the classical macrodispersion theory may not be applicable. The methodology relies on the observation that breakthrough curves in heterogeneous media are generally well described by lognormal distributions, and mean breakthrough times can be predicted analytically. The log-variance of solute arrival is thus sufficient to completely specify the breakthrough curves, and this is calibrated as a function of aquifer heterogeneity and dimensionless distance from a source plane by means of Monte Carlo analysis and statistical regression. Using the ensemble of simulated groundwater flow and solute transport realizations employed to calibrate the predictive regression, reliability estimates for the prediction are also developed. Additional theoretical contributions include heuristics for the time until an effective macrodispersion coefficient becomes applicable, and also an expression for its magnitude that applies in highly heterogeneous systems. Finally, it is seen that the results here represent a way to derive continuous time random walk transition distributions from physical considerations rather than from empirical field calibration.

58 GEOSCIENCES↗

DOE OSTI2021

Compositional effects on cure kinetics, mechanical properties and printability of dual-cure epoxy/acrylate resins for DIW additive manufacturing

Interest in 3D printing of thermoset resins has increased significantly in recent years. One approach to additive manufacturing of thermoset resins is printing dual-cure resins with direct ink write (DIW). Dual-cure resins are multi-component resins which employ an in situ curable constituent to enable net-shape fabrication while a second constituent and cure mechanism contribute to the final mechanical properties of the printed materials. In this work, the cure kinetics, green strength, printability, and print fidelity of dual-cure epoxy/acrylate thermoset resins are investigated. Resin properties are evaluated as a function of acrylate concentration and in situ UV exposure conditions. The acrylate cure kinetics are probed using photo-differential scanning calorimetry and the impacts of resin composition and UV cure profile on the acrylate extent of conversion are presented. Continuous and pulsed UV cure profiles are shown to affect total conversion due to variances in radical efficiency at different UV intensities and acrylate concentrations. The effects of acrylate concentration on the kinetics of the epoxy thermal cure and the final mechanical properties are also investigated using dynamic mechanical analysis and three-point bend measurements. The glass transition temperature is dependent on formulation, with increasing acrylate content decreasing the T g . However, the room temperature shear moduli, flexural moduli, strength, strain-to-failure, and toughness values are relatively independent of resin composition. The similarity of the final properties allows for greater flexibility in resin formulation and in situ cure parameters, which can enable the printing of complex parts that require high green strength. We found that the in situ UV print intensities and exposure profiles that are necessary to achieve the best print quality are not, in most cases, the conditions that maximize conversion of the acrylate network. Lastly, this highlights the importance of developing optimized resin compositions which enable complete cure of the acrylate network by promoting acrylate dark cure or thermal cure.

36 MATERIALS SCIENCE↗

DOE OSTI2014

On the estimation of statistical uncertainties on powder diffraction and small-angle scattering data from two-dimensional X-ray detectors

Optimal methods are explored for obtaining one-dimensional powder pattern intensities from two-dimensional planar detectors with good estimates of their standard deviations. Methods are described to estimate uncertainties when the same image is measured in multiple frames as well as from a single frame. The importance of considering the correlation of diffraction points during the integration and the resampling process of data analysis is shown. It is found that correlations between adjacent pixels in the image can lead to seriously overestimated uncertainties if such correlations are neglected in the integration process. Off-diagonal entries in the variance–covariance (VC) matrix are problematic as virtually all data processing and modeling programs cannot handle the full VC matrix. It is shown that the off-diagonal terms come mainly from the pixel-splitting algorithm used as the default integration algorithm in many popular two-dimensional integration programs, as well as from rebinning and resampling steps later in the processing. When the full VC matrix can be propagated during the data reduction, it is possible to get accurate refined parameters and their uncertainties at the cost of increasing computational complexity. However, as this is not normally possible, the best approximate methods for data processing in order to estimate uncertainties on refined parameters with the greatest accuracy from just the diagonal variance terms in the VC matrix is explored.

Yang, X.↗

DOE OSTI2024

Improving the estimate of higher-order moments from lidar observations near the top of the convective boundary layer

Abstract. Ground-based lidar data have proven extremely useful for profiling the convective boundary layer (CBL). Many groups have derived higher-order moments (e.g., variance, skewness, fluxes) from high-temporal-resolution lidar data using an autocovariance approach. However, these analyses are highly uncertain near the CBL top when the depth of the CBL (zi) is changing during the analysis period. This is because the autocovariance approach is usually applied to constant height levels and the character of the eddies is changing on either side of the changing CBL top. Here, a new approach is presented wherein the autocovariance analysis is performed on a normalized height grid, with a temporally smoothed zi. Output from a large eddy simulation model demonstrates that deriving higher-order moments from time series on a normalized height grid has better agreement with the slab-averaged quantities than the moments derived from the original height grid.

Rosenberger, Tessa E. (ORCID:0000000333205873)↗

DOE OSTI2023

BigNeuron: a resource to benchmark and predict performance of algorithms for automated tracing of neurons in light microscopy datasets

BigNeuron is an open community bench-testing platform with the goal of setting open standards for accurate and fast automatic neuron tracing. We gathered a diverse set of image volumes across several species that is representative of the data obtained in many neuroscience laboratories interested in neuron tracing. Here, we report generated gold standard manual annotations for a subset of the available imaging datasets and quantified tracing quality for 35 automatic tracing algorithms. The goal of generating such a hand-curated diverse dataset is to advance the development of tracing algorithms and enable generalizable benchmarking. Together with image quality features, we pooled the data in an interactive web application that enables users and developers to perform principal component analysis, t-distributed stochastic neighbor embedding, correlation and clustering, visualization of imaging and tracing data, and benchmarking of automatic tracing algorithms in user-defined data subsets. The image quality metrics explain most of the variance in the data, followed by neuromorphological features related to neuron size. Furthermore, we observed that diverse algorithms can provide complementary information to obtain accurate results and developed a method to iteratively combine methods and generate consensus reconstructions. The consensus trees obtained provide estimates of the neuron structure ground truth that typically outperform single algorithms in noisy datasets. However, specific algorithms may outperform the consensus tree strategy in specific imaging conditions. Finally, to aid users in predicting the most accurate automatic tracing results without manual annotations for comparison, we used support vector machine regression to predict reconstruction quality given an image volume and a set of automatic tracings.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2015

Assessing the use of reflectance spectroscopy in determining CsCl stress in the model species Arabidopsis thaliana

Here, reflectance spectroscopy is a rapid and non-destructive analytical technique that may be used for assessing plant stress, and has potential applications for use in remediation. Changes in reflectance such as that due to metal stress may occur before damage is visible, and existing studies have shown that metal stress does cause changes in plant reflectance. To further investigate the potential use of reflectance spectroscopy as a method for assessing metal stress in plants, an exploratory study was conducted in which Arabidopsis thaliana plants were treated twice weekly in a laboratory setting with varying levels (0, 0.5, or 5 mM (millimolar)) of caesium chloride (CsCl) solution, and reflectance spectra were collected every week for three weeks using an Analytical Spectral Devices FieldSpec Pro spectroradiometer with both a contact probe (CP) and a field of view (FOV) probe at 36.8 and 66.7 cm, respectively, above the plant. Plants were harvested each week after spectra collection for determination of relative water content and chlorophyll content. A visual assessment of the plants was also conducted using point observations on a uniform grid of 81 points. A mixed-effects model analysis was conducted for each vegetation index (VI) considered to determine the effects of length of treatment, treatment level, view with which spectra were acquired, and the interactions of these terms. Two-way analyses of variance (ANOVAs) were performed on the aforementioned endpoints (e.g. chlorophyll content) to determine the significance of the effects of treatment level and length of treatment. Multiple linear regression (MLR) was used to develop a predictive model for each endpoint, considering VI acquired at each view (CP, high FOV, and low FOV). Of the 14 VI considered, 8 were included in the MLR models. Contact probe readings and FOV readings differed significantly, but FOV measurements were generally consistent at each height.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2022

Machine Learning of All Mycobacterium tuberculosis H37Rv RNA-seq Data Reveals a Structured Interplay between Metabolism, Stress Response, and Infection

Mycobacterium tuberculosis is one of the most consequential human bacterial pathogens, posing a serious challenge to 21st century medicine. A key feature of its pathogenicity is its ability to adapt its transcriptional response to environmental stresses through its transcriptional regulatory network (TRN). While many studies have sought to characterize specific portions of the M. tuberculosis TRN, and some studies have performed system-level analysis, few have been able to provide a network-based model of the TRN that also provides the relative shifts in transcriptional regulator activity triggered by changing environments. Here, we compiled a compendium of nearly 650 publicly available, high quality M. tuberculosis RNA-sequencing data sets and applied an unsupervised machine learning method to obtain a quantitative, top-down TRN. It consists of 80 independently modulated gene sets known as “iModulons,” 41 of which correspond to known regulons. These iModulons explain 61% of the variance in the organism’s transcriptional response. We show that iModulons (i) reveal the function of poorly characterized regulons, (ii) describe the transcriptional shifts that occur during environmental changes such as shifting carbon sources, oxidative stress, and infection events, and (iii) identify intrinsic clusters of regulons that link several important metabolic systems, including lipid, cholesterol, and sulfur metabolism. This transcriptome-wide analysis of the M. tuberculosis TRN informs future research on effective ways to study and manipulate its transcriptional regulation and presents a knowledge-enhanced database of all published high-quality RNA-seq data for this organism to date.

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2023

Driving mode analysis—How uncertain functional inputs propagate to an output

Abstract Driving mode analysis elucidates how correlated features of uncertain functional inputs jointly propagate to produce uncertainty in the output of a computation. Uncertain input functions are decomposed into three terms: the mean functions, a zero‐mean driving mode, and zero‐mean residual. The random driving mode varies along a single direction, having fixed functional shape and random scale. It is uncorrelated with the residual, and under linear error propagation, it produces an output variance equal to that of the full input uncertainty. Finally, the driving mode best represents how input uncertainties propagate to the output because it minimizes expected squared Mahalanobis distance amongst competitors. These characteristics recommend interpretation of the driving mode as the single‐degree‐of‐freedom component of input uncertainty that drives output uncertainty. We derive the functional driving mode, show its superiority to other seemingly sensible definitions, and demonstrate the utility of driving mode analysis in an application. The application is the simulation of neutron transport in criticality experiments. The uncertain input functions are nuclear data that describe how Pu reacts to bombardment by neutrons. Visualization of the driving mode helps scientists understand what aspects of correlated functional uncertainty have effects that either reinforce or cancel one another in propagating to the output of the simulation.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2016

Planck 2015 results: XX. Constraints on inflation

In this paper, we present the implications for cosmic inflation of the Planck measurements of the cosmic microwave background (CMB) anisotropies in both temperature and polarization based on the full Planck survey, which includes more than twice the integration time of the nominal survey used for the 2013 release papers. The Planck full mission temperature data and a first release of polarization data on large angular scales measure the spectral index of curvature perturbations to be n s = 0.968 ± 0.006 and tightly constrain its scale dependence to dn s / dlnk = -0.003 ± 0.007 when combined with the Planck lensing likelihood. When the Planck high-ℓ polarization data are included, the results are consistent and uncertainties are further reduced. The upper bound on the tensor-to-scalar ratio is r 0.002 < 0.11 (95% CL). This upper limit is consistent with the B-mode polarization constraint r< 0.12 (95% CL) obtained from a joint analysis of the BICEP2/Keck Array and Planck data. These results imply that V(φ) ∝ φ 2 and natural inflation are now disfavoured compared to models predicting a smaller tensor-to-scalar ratio, such as R 2 inflation. We search for several physically motivated deviations from a simple power-law spectrum of curvature perturbations, including those motivated by a reconstruction of the inflaton potential not relying on the slow-roll approximation. We find that such models are not preferred, either according to a Bayesian model comparison or according to a frequentist simulation-based analysis. Three independent methods reconstructing the primordial power spectrum consistently recover a featureless and smooth P R (k)over the range of scales 0.008 Mpc -1 ≲ k ≲ 0.1 Mpc -1 . At large scales, each method finds deviations from a power law, connected to a deficit at multipoles ℓ ≈ 20-40 in the temperature power spectrum, but at an uncompelling statistical significance owing to the large cosmic variance present at these multipoles. By combining power spectrum and non-Gaussianity bounds, we constrain models with generalized Lagrangians, including Galileon models and axion monodromy models. The Planck data are consistent with adiabatic primordial perturbations, and the estimated values for the parameters of the base Λ cold dark matter (ΛCDM) model are not significantly altered when more general initial conditions are admitted. In correlated mixed adiabatic and isocurvature models, the 95% CL upper bound for the non-adiabatic contribution to the observed CMB temperature variance is | α non - adi | < 1.9%, 4.0%, and 2.9% for CDM, neutrino density, and neutrino velocity isocurvature modes, respectively. We have tested inflationary models producing an anisotropic modulation of the primordial curvature power spectrum finding that the dipolar modulation in the CMB temperature field induced by a CDM isocurvature perturbation is not preferred at a statistically significant level. We also establish tight constraints on a possible quadrupolar modulation of the curvature perturbation. Lastly, these results are consistent with the Planck 2013 analysis based on the nominal mission data and further constrain slow-roll single-field inflationary models, as expected from the increased precision of Planck data using the full set of observations.

79 ASTRONOMY AND ASTROPHYSICS↗

DOE OSTI2013

In-Situ Statistical Analysis of Autotune Simulation Data using Graphical Processing Units

Developing accurate building energy simulation models to assist energy efficiency at speed and scale is one of the research goals of the Whole-Building and Community Integration group, which is a part of Building Technologies Research and Integration Center (BTRIC) at Oak Ridge National Laboratory (ORNL). The aim of the Autotune project is to speed up the automated calibration of building energy models to match measured utility or sensor data. The workflow of this project takes input parameters and runs EnergyPlus simulations on Oak Ridge Leadership Computing Facility’s (OLCF) computing resources such as Titan, the world’s second fastest supercomputer. Multiple simulations run in parallel on nodes having 16 processors each and a Graphics Processing Unit (GPU). Each node produces a 5.7 GB output file comprising 256 files from 64 simulations. Four types of output data covering monthly, daily, hourly, and 15- minute time steps for each annual simulation is produced. A total of 270TB+ of data has been produced. In this project, the simulation data is statistically analyzed in-situ using GPUs while annual simulations are being computed on the traditional processors. Titan, with its recent addition of 18,688 Compute Unified Device Architecture (CUDA) capable NVIDIA GPUs, has greatly extended its capability for massively parallel data processing. CUDA is used along with C/MPI to calculate statistical metrics such as sum, mean, variance, and standard deviation leveraging GPU acceleration. The workflow developed in this project produces statistical summaries of the data which reduces by multiple orders of magnitude the time and amount of data that needs to be stored. These statistical capabilities are anticipated to be useful for sensitivity analysis of EnergyPlus simulations

accelerators↗

DOE OSTI2016

SU-F-R-34: Quantitative Perfusion Measurement in Rectal Cancer Using Three Different Pharmacokinetic Models: Implications for Prospective Study Design

Purpose: To compare three different pharmacokinetic models for analysis of dynamic-contrast-enhanced (DCE)-CT data with respect to different acquisition times and location of region of interest. Methods: Eight rectal cancer patients with pre-treatment DCE-CTs were included. The dynamic sequence started 4–10seconds(s) after the injection of contrast agent. The scan included a 110s acquisition with intervals of 40×1s+15×3s+4×6s. An experienced oncologist outlined the tumor region. Hotspots with top-5%-enhancement were also identified. Pharmacokinetic analysis was performed using three different models: deconvolution method, Patlak model, and modified Toft’s model. Perfusion parameters as blood flow (BF), blood volume (BV), mean transit time (MTT), permeability-surface-area-product (PS), volume transfer constant (Ktrans), and flux rate constant (Kep), were compared with respect to different acquisition times of 45s, 65s, 85s and 105s. Both hotspot and whole-volume variances were also assessed. The differences were compared using the Wilcoxon matched-pairs test and Bland-Altman plots. Results: Moderate correlation was observed for various perfusion parameters (r=0.56–0.72, p<0.0001) but the Wilcoxon test revealed a significant difference among the three models (P < .001). Significant differences in PS were noted between acquisitions of 45s versus longer time of 85s or 105s (p<0.05) using Patlak but not with the deconvolution method. In addition, measurements varied substantially between whole-volume vs. hotspot analysis. Conclusion: The radiation dose of DCE-CT was on average 1.5 times of an abdomen/pelvic CT, which is not insubstantial. To take the DCE-CT forward as a biomarker in oncology, prospective studies should be carefully designed with the optimal image acquisition and analysis technique. Our study suggested that: (1) different kinetic models are not interchangeable; (2) a 45s acquisition might not be sufficient for reliable permeability measurement in rectal cancer using Patlak model, but might be achievable using deconvolution method; and (3) local variations existed inside the tumor, and both whole-volume-averaged and local-heterogeneity analysis is recommended for future quantitative studies. This work is supported by the National High-tech R&D program for Young Scientists by the Ministry of Science and Technology of China (Grant No. 2015AA020917), Natural Science Foundation of China (NSFC Grant No. 81201091).

60 APPLIED LIFE SCIENCES↗