Search NASA⌕ Search

SEARCH · Search NASA

Results for “reduced set methods”

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 415 records · Page 23

Computation of the spectrum of spatial Lyapunov exponents for the spatially extended beam-plasma systems and electron-wave devices

The spectrum of Lyapunov exponents is powerful tool for the analysis of the complex system dynamics. In the general framework of nonlinear dynamics, a number of the numerical techniques have been developed to obtain the spectrum of Lyapunov exponents for the complex temporal behavior of the systems with a few degree of freedom. Unfortunately, these methods cannot be applied directly to analysis of complex spatio-temporal dynamics of plasma devices which are characterized by the infinite phase space, since they are the spatially extended active media. In the present paper, we propose the method for the calculation of the spectrum of the spatial Lyapunov exponents (SLEs) for the spatially extended beam-plasma systems. The calculation technique is applied to the analysis of chaotic spatio-temporal oscillations in three different beam-plasma model: (1) simple plasma Pierce diode, (2) coupled Pierce diodes, and (3) electron-wave system with backward electromagnetic wave. We find an excellent agreement between the system dynamics and the behavior of the spectrum of the spatial Lyapunov exponents. Along with the proposed method, the possible problems of SLEs calculation are also discussed. It is shown that for the wide class of the spatially extended systems, the set of quantities included in the system state for SLEs calculation can be reduced using the appropriate feature of the plasma systems.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

A steepest descents method for reentry optimization

A steepest descents optimization program is applied to the problem of a lifting vehicle entering the earth's atmosphere. The program employs penalty functions representing terminal conditions and inflight inequality constraints. During each iteration, it reduces a single performance measure which is the sum of the performance index and the penalty functions. Therefore, only one set of adjoint equations must be integrated per iteration. Values of weight factors, multiplying the penalty functions, are automatically adjusted before each iteration in order that the penalty functions will approach acceptable values. This method is shown to be a form of the classical Lagrange multiplier methods.

Andrus, J. F.↗

Effectively using multifidelity optimization for wind turbine design

Abstract. Wind turbines are complex multidisciplinary systems that are challenging to design because of the tightly coupled interactions between different subsystems. Computational modeling attempts to resolve these couplings so we can efficiently explore new wind turbine systems early in the design process. Low-fidelity models are computationally efficient but make assumptions and simplifications that limit the accuracy of design studies, whereas high-fidelity models capture more of the actual physics but with increased computational cost. This paper details the use of multifidelity methods for optimizing wind turbine designs by using information from both low- and high-fidelity models to find an optimal solution at reduced cost. Specifically, a trust-region approach is used with a novel corrective function built from a nonlinear surrogate model. We find that for a diverse set of design problems – with examples given in rotor blade geometry design, wind turbine controller design, and wind power plant layout optimization – the multifidelity method finds the optimal design using 38 %–58 % of the computational cost of the high-fidelity-only optimization. The success of the multifidelity method in disparate applications suggests that it could be more broadly applied to other wind energy or otherwise generic applications.

17 WIND ENERGY↗

Iterative self-organizing SCEne-LEvel sampling (ISOSCELES) for large-scale building extraction

Convolutional neural networks (CNN) provide state-of-the-art performance in many computer vision tasks, including those related to remote-sensing image analysis. Successfully training a CNN to generalize well to unseen data, however, requires training on samples that represent the full distribution of variation of both the target classes and their surrounding contexts. With remote sensing data, acquiring a sufficiently representative training set is a challenge due to both the inherent multi-modal variability of satellite or aerial imagery and the general high cost of labeling data. To address this challenge, we have developed ISOSCELES, an Iterative Self-Organizing SCEne LEvel Sampling method for hierarchical sampling of large image sets. Using affinity propagation, ISOSCELES automates the selection of highly representative training images. Compared to random sampling or using available reference data, the distribution of the training is principally data driven, reducing the chance of oversampling uninformative areas or undersampling informative ones. In comparison to manual sample selection by an analyst, ISOSCELES exploits descriptive features, spectral and/or textural, and eliminates human bias in sample selection. Using a hierarchical sampling approach, ISOSCELES can obtain a training set that reflects both between-scene variability, such as in viewing angle and time of day, and within-scene variability at the level of individual training samples. We verify the method by demonstrating its superiority to stratified random sampling in the challenging task of adapting a pre-trained model to a new image and spatial domain for country-scale building extraction. Using a pair of hand-labeled training sets comprising 1,987 sample image chips, a total of 496,000,000 individually labeled pixels, we show, across three distinct model architectures, an increase in accuracy, as measured by F1-score, of 2.2–4.2%.

42 ENGINEERING↗

Recent progress in the design of the K-DEMO divertor

The preliminary conceptual design of the Korean fusion demonstration reactor (K-DEMO) with a major radius of 6.8 m and the fusion power of 2200 MW has been studied since 2012. The overall configuration of the K-DEMO divetor system based on the ITER-like water-cooled tungsten technology is a double-null type symmetric divertor subdivided into 32 toroidal modules for the vertical maintenance. A detached divertor scenario with impurity seeding was considered as the primary approach for the power exhaust to reduce the peak heat flux lower than the engineering limit of 10 MW/m 2 . The power exhaust performance at the scrap off layer was estimated by using UEDGE-2D code, a two-dimensional fluid transport code for collisional edge plasma and neutral species like N, Ne, and Ar. Particle and heat flux on inboard and outboard divertor targets were calculated for the detached cased depending on parameters such as the impurity seeding rate, pumping rate, and the pedestal density. On the other hand, a magnetic solution like X-divertor, snowflake divertor, and super X-divertor to expand the plasma wet area was considered for K-DEMO since the detached divertor increasing a radiation fraction by impurity seeding might be able to be unstable. However, the extremely high current of poloidal coils was required more than the engineering limit, 20 MA, to form magnetic field lines for the alternative divertors. Based on the physical calculation of the edge plasma, engineering analyses were carried out to find out the thermal and structural reliability. The thermo-hydraulic analysis confirmed thermal stability, whether all comprising materials are operating within their allowable temperature windows when the case of the peak heat flux is set to 10 MW/m 2 on the outboard divertor target. The response surface optimization method derived two optimal design candidates employing two kinds of heat sink materials, respectively: the reduced activation ferritic martensitic (RAFM) steel and CuCrZr alloy. Additionally, the drawbacks and merits of the two materials were definite. The optimal design with applying RAFM steel was vulnerable to withstand thermal and mechanical loads since low thermal conductivity caused too thin thickness of the heat sink. On the other hand, the CuCrZr alloy has critical drawbacks in terms of activation and radioactive waste despite its high thermal conductivity. Meanwhile, preliminary electromagnetic (EM) analysis was carried out to estimate the EM loads caused by the abnormal behaviors of plasma since EM loads are one of the most critical external loads for designing a DEMO divertor.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Waste reduction process improvements in the analysis of plutonium by x-ray fluorescence: results from multiple data sets

To minimize waste, improve process safety, and minimize costs, modifications were implemented to a method for quantifying gallium in plutonium metal using wavelength dispersive X-ray fluorescence. These changes included reducing sample sizes, reducing ion exchange process volumes, using cheaper reagent grade acids, eliminating the use of HF acid, and using more robust containment film for sample analysis. Relative precision and accuracy achieved from analyzing multiple aliquots from a single parent sample were {approx}0.2% and {approx}0.1% respectively. The same precision was obtained from analyzing a total of four parent materials, and the average relative accuracy from all the samples was 0.4%, which is within programmatic uncertainty requirements.

36 MATERIALS SCIENCE↗

Method of up-front load balancing for local memory parallel processors

In a parallel processing computer system with multiple processing units and shared memory, a method is disclosed for uniformly balancing the aggregate computational load in, and utilizing minimal memory by, a network having identical computations to be executed at each connection therein. Read-only and read-write memory are subdivided into a plurality of process sets, which function like artificial processing units. Said plurality of process sets is iteratively merged and reduced to the number of processing units without exceeding the balance load. Said merger is based upon the value of a partition threshold, which is a measure of the memory utilization. The turnaround time and memory savings of the instant method are functions of the number of processing units available and the number of partitions into which the memory is subdivided. Typical results of the preferred embodiment yielded memory savings of from sixty to seventy five percent.

Baffes, Paul Thomas↗

The POINTER Imaging baseline cohort: Associations between multimodal neuroimaging biomarkers, cardiovascular health, and cognition

Abstract INTRODUCTION The U.S. Study to Protect Brain Health Through Lifestyle Intervention to Reduce Risk (U.S. POINTER) is evaluating lifestyle interventions in older adults at risk for cognitive decline and dementia. Here we characterize the baseline data set of the POINTER Imaging ancillary study. METHODS Participants underwent health and cognitive assessments and neuroimaging with multimodal positron emission tomography (PET) (beta‐amyloid [Aβ] and tau) and magnetic resonance imaging (MRI). Framingham risk score (FRS) was used to quantify cardiovascular disease (CVD) risk. RESULTS A total of 1052 participants (31% from underrepresented ethnoracial groups) were enrolled. Compared to Aβ−, Aβ+ (29%) participants were older, had higher apolipoprotein E (APOE) ε4 carriage rate and white matter hyperintensity volume, and greater temporal tau. FRS was related to MRI measures, but not AD biomarkers. FRS and tau had independent effects on cognition. DISCUSSION In this heterogenous, at‐risk cohort, CVD risk was related to more abnormal brain structure and poorer cognition, representing a putative non‐AD (Alzheimer's disease) pathway to brain injury and cognitive decline. Highlights The U.S. Study to Protect Brain Health Through Lifestyle Intervention to Reduce Risk (U.S. POINTER) cohort is enriched for cardiovascular disease (CVD) and poor lifestyle POINTER Imaging collected multimodal neuroimaging data in this unique, at‐risk cohort Amyloid burden was related to age, apolipoprotein E (APOE) ε4 carriage, and measures of disease progression Associations between amyloid and tau, and tau and cognition, were relatively weak CVD risk and tau pathology were independently related to memory

Neurosciences & Neurology↗

Correlation of finite element free vibration predictions using random vibration test data

Finite element analysis is regularly used during the engineering cycle of mechanical systems to predict the response to static, thermal, and dynamic loads. The finite element model (FEM) used to represent the system is often correlated with physical test results to determine the validity of analytical results provided. Results from dynamic testing provide one means for performing this correlation. One of the most common methods of measuring accuracy is by classical modal testing, whereby vibratory mode shapes are compared to mode shapes provided by finite element analysis. The degree of correlation between the test and analytical mode shapes can be shown mathematically using the cross orthogonality check. A great deal of time and effort can be exhausted in generating the set of test acquired mode shapes needed for the cross orthogonality check. In most situations response data from vibration tests are digitally processed to generate the mode shapes from a combination of modal parameters, forcing functions, and recorded response data. An alternate method is proposed in which the same correlation of analytical and test acquired mode shapes can be achieved without conducting the modal survey. Instead a procedure is detailed in which a minimum of test information, specifically the acceleration response data from a random vibration test, is used to generate a set of equivalent local accelerations to be applied to the reduced analytical model at discrete points corresponding to the test measurement locations. The static solution of the analytical model then produces a set of deformations that once normalized can be used to represent the test acquired mode shapes in the cross orthogonality relation. The method proposed has been shown to provide accurate results for both a simple analytical model as well as a complex space flight structure.

Chambers, Jeffrey A.↗

On-the-fly response function generation method for composite coarse mesh

The hybrid stochastic deterministic transport code COMET, based on the incident response expansion theory, is used to model reactor cores with high fidelity and formidable computational speed. COMET models a reactor core using a library of incident flux response expansion coefficients that are pre computed for all the unique lattice cells (e.g., fuel assemblies, reflector blocks, etc.) in the core. In order to further improve its computational efficiency in pre-calculating the response library a new response function generation method is developed to compute the response functions for the composite coarse meshes made of a smaller set of unique lattices on the fly within the COMET's deterministic transport core sweep. The efficiency is achieved by eliminating a number of unique lattices that can be made up from the reduced set of unique meshes on the fly. The numerical process consists of the following steps. First, the boundary condition on composite coarse mesh boundaries is projected onto the expansion basis to compute the incident flux moments on external surfaces of all the basic (reduced set of unique) coarse meshes. Secondly, the deterministic sweeping solver in COMET is used to converge on the outgoing/incoming flux expansion moments crossing interfaces between the basic coarse meshes. Thirdly, the response functions for the composite coarse meshes are constructed as a superposition on the fly. The new response function generation method was tested on 88 composite coarse meshes consisting of CANDU fuel bundles and moderator blocks. It was found that response functions generated by the new method agree very well with those generated by direct Monte Carlo calculations. The average and maximum relative differences in the surface-to-surface response coefficients computed by the two methods are 0.10% and 0.20%, respectively. Similarly, the average and maximum relative differences in the response fission densities are 0.13% and 0.43%, respectively. These discrepancies are within one standard deviation of the stochastic uncertainties. The new method is five times faster than the original direct Monte Carlo method. The size of the response function library for the new method is five times smaller than that for the original method, leading to significantly less requirement for the computer hard drive space and memory. (authors)

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

Active operator learning with predictive uncertainty quantification for partial differential equations

With the increased prevalence of neural operators being used to provide rapid solutions to partial differential equations (PDEs), understanding the accuracy of model predictions and the associated error levels is necessary for deploying reliable surrogate models in scientific applications. Existing uncertainty quantification (UQ) frameworks employ ensembles or Bayesian methods, which can incur substantial computational costs during both training and inference. Here, we propose a lightweight predictive UQ method tailored for Deep operator networks (DeepONets) that also generalizes to other operator networks. Numerical experiments on linear and nonlinear PDEs demonstrate that the framework’s uncertainty estimates are unbiased and provide accurate out-of-distribution uncertainty predictions with a sufficiently large training dataset. Our framework provides fast inference and uncertainty estimates that can efficiently drive outer-loop analyses that would be prohibitively expensive with conventional solvers. We demonstrate how predictive uncertainties can be used in the context of Bayesian optimization and active learning problems to yield improvements in accuracy and data-efficiency for outer-loop optimization procedures. In the active learning setup, we extend the framework to Fourier Neural Operators (FNO) and describe a generalized method for other operator networks. To enable real-time deployment, we introduce an inference strategy based on precomputed trunk outputs and a sparse placement matrix, reducing evaluation time by more than a factor of five. Our method provides a practical route to uncertainty-aware operator learning in time-sensitive settings.

97 MATHEMATICS AND COMPUTING↗

Coarse grained open system quantum dynamics

We show that the quantum dynamics of a system comprised of a subspace Q coupled to a larger subspace P can be recast as a reduced set of 'coarse grained' ordinary differential equations with constant coefficients. These equations can be solved by a single diagonalization of a general complex matrix. The method makes no assumptions about the strength of the couplings between the Q and the P subspaces, nor is there any limitation on the initial population in P. The utility of the method is demonstrated via computations in three following areas: molecular compounds, photonic materials, and condensed phases.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Iterative diagonalization of the non-Hermitian transcorrelated Hamiltonian using a plane-wave basis set: Application to sp-electron systems with deep core states

We develop an iterative diagonalization scheme in solving a one-body self-consistent-field equation in the transcorrelated (TC) method using a plane-wave basis set. Non-Hermiticity in the TC method is well handled with a block-Davidson algorithm. We verify that the required computational cost is efficiently reduced by our algorithm. In addition, we apply our plane-wave-basis TC calculation to some simple sp-electron systems with deep core states to elucidate an impact of the pseudopotential approximation to the calculated band structures. We find that a position of the deep valence bands is improved by an explicit inclusion of core states, but an overall band structure is consistent with a regular setup that includes core states into the pseudopotentials. This study offers an important understanding for the future application of the TC method to strongly correlated solids.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Duality-Free Decomposition Based Data-Driven Stochastic Security-Constrained Unit Commitment

To incorporate the superiority of both stochastic and robust approaches, a data-driven stochastic optimization is employed to solve the security-constrained unit commitment model. This approach makes the most use of the historical data to generate a set of possible probability distributions for wind power outputs and then it optimizes the unit commitment under the worst-case probability distribution. However, this model suffers from huge computational burden, as a large number of scenarios are considered. To tackle this issue, a duality-free decomposition method is proposed in this paper. This approach does not require doing duality, which can save a large set of dual variables and constraints, and therefore reduces the computational burden. In addition, the inner max-min problem has a special mathematical structure, where the scenarios have the similar constraint. Thus, the max-min problem can be decomposed into independent sub-problems to be solved in parallel, which further improves the computational efficiency. A numerical study on an IEEE 118-bus system with practical data of a wind power system has demonstrated the effectiveness of the proposal.

42 ENGINEERING↗

SU-F-J-181: An Alternative Patient Alignment Tool On TomoTherapy: The First In- Human Megavoltage-Topogram Acquisition

Purpose: To show the first in-human Megavoltage (MV)-Topogram acquisition for the evaluation of the potential for MV-Topogram-based alignment as an alternative to MVCT for reducing dose and imaging time. Methods: A lung cancer patient was enrolled in an ongoing IRB-approved clinical trial at our institute. The patient was set up using the clinical protocol employing positioning lasers. 3.2mm diameter tungsten spheres were placed on the patient’s skin at their alignment tattoos to check surface-based marker concordance between topograms and MVCT. Anterior-Posterior (AP) and lateral (LAT) MV-Topograms were acquired using gantry angles of 0°/90° with a 1mm collimator opening, all MLC leafs open, 4cm/s couch speed, and 12.5s scanning time. The topogram acquisition was immediately followed by the normal MVCT scan acquisition. MV-Topograms were reconstructed from the detector exit-data using in-house developed software. The topograms were also enhanced using contrast-limited adaptive histogram equalization (CLAHE). The MV-Topograms were registered to reference kV-based digitally reconstructed topograms. The localization results were compared against results obtained comparing the clinical MVCT to the kVCT simulation. Results: The shifts using the unenhanced Topograms, enhanced Topograms, and MVCT were (LAT, LONG, VERT, ROLL) (5.8mm, 2.6mm, −5.6mm, 0.34°), (3.9mm, 2.5mm, −2.2mm, 0.65°) and (2.4mm, 1.5mm, −3.0mm, 0.5°), respectively. The magnitude alignment differences between the enhanced Topograms and MVCT were within 1.5 mm and 0.15°. The average MVCT and total Topogram acquisition times were 272.9s ± 31.5s and 46s, respectively. Conclusion: MV-Topograms have the potential for providing equivalent performance with less dose and acquisition time than the traditional MVCT technique. We are evaluating other sites as well as adding patients to develop statistically significant analyses regarding the alignment quality differences. MV-Topograms are likely to be most clinically useful for bony anatomy and radiopaque marker-based alignments. The study was supported by an Accuray Grant.

60 APPLIED LIFE SCIENCES↗

Manifold Learning: What, How, and Why

Manifold learning (ML), also known as nonlinear dimension reduction, is a set of methods to find the low-dimensional structure of data. Dimension reduction for large, high-dimensional data is not merely a way to reduce the data; the new representations and descriptors obtained by ML reveal the geometric shape of high-dimensional point clouds and allow one to visualize, denoise, and interpret them. This review presents the underlying principles of ML, its representative methods, and their statistical foundations, all from a practicing statistician's perspective. It describes the trade-offs and what theory tells us about the parameter and algorithmic choices we make in order to obtain reliable conclusions.

Mathematics↗

Simulating Welding-Robot Trajectories For Previewing

New method devised for generating welding-tool paths; parts of off-line-programmed test trajectories of computer-controlled welding robot. Reduces time necessary to set up simulation of welding process and eliminates some errors by reducing amount of repetition. Provides visual feedback similar to graphical simulation used in off-line-programming process after development of model.

Levitt, Maureen L.↗

An accelerated training method for back propagation networks

The principal objective is to provide a training procedure for a feed forward, back propagation neural network which greatly accelerates the training process. A set of orthogonal singular vectors are determined from the input matrix such that the standard deviations of the projections of the input vectors along these singular vectors, as a set, are substantially maximized, thus providing an optimal means of presenting the input data. Novelty exists in the method of extracting from the set of input data, a set of features which can serve to represent the input data in a simplified manner, thus greatly reducing the time/expense to training the system.

Shelton, Robert O.↗