Search NASA⌕ Search

SEARCH · Search NASA

Results for “kernel method”

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 1,027 records · Page 57

Cross-correlation of Planck CMB lensing with DESI-like LRGs

ABSTRACT Cross-correlations between the lensing of the cosmic microwave background (CMB) and other tracers of large-scale structure provide a unique way to reconstruct the growth of dark matter, break degeneracies between cosmology and galaxy physics, and test theories of modified gravity. We detect a cross-correlation between Dark Energy Spectroscopic Instrument (DESI)-like luminous red galaxies (LRGs) selected from DECam Legacy Survey imaging and CMB lensing maps reconstructed with the Planck satellite at a significance of S/N = 27.2 over scales ℓmin = 30, ℓmax = 1000. To correct for magnification bias, we determine the slope of the LRG cumulative magnitude function at the faint limit as s = 0.999 ± 0.015, and find corresponding corrections of the order of a few per cent for $C^{\kappa g}_{\ell }, C^{gg}_{\ell }$ across the scales of interest. We fit the large-scale galaxy bias at the effective redshift of the cross-correlation zeff ≈ 0.68 using two different bias evolution agnostic models: a HaloFit times linear bias model where the bias evolution is folded into the clustering-based estimation of the redshift kernel, and a Lagrangian perturbation theory model of the clustering evaluated at zeff. We also determine the error on the bias from uncertainty in the redshift distribution; within this error, the two methods show excellent agreement with each other and with DESI survey expectations.

79 ASTRONOMY AND ASTROPHYSICS↗

Development of advanced machine learning models for analysis of plutonium surrogate optical emission spectra

This work investigates and applies machine learning paradigms seldom seen in analytical spectroscopy for quantification of gallium in cerium matrices via processing of laser-plasma spectra. Ensemble regressions, support vector machine regressions, Gaussian kernel regressions, and artificial neural network techniques are trained and tested on cerium-gallium pellet spectra. A thorough hyperparameter optimization experiment is conducted initially to determine the best design features for each model. The optimized models are evaluated for sensitivity and precision using the limit of detection (LoD) and root mean-squared error of prediction (RMSEP) metrics, respectively. Gaussian kernel regression yields the superlative predictive model with an RMSEP of 0.33% and an LoD of 0.015% for quantification of Ga in a Ce matrix. This study concludes that these machine learning methods could yield robust prediction models for rapid quality control analysis of plutonium alloys.

Rao, Ashwin P. (ORCID:0000000319312568)↗

Exploring the Random Phase Approximation for Materials and Chemical Physics

In this work we test the performance of the fully nonlocal random phase approximation (RPA) for challenging problems in materials chemistry and physics. In this work we assessed the fully nonlocal random phase approximation (RPA) for the delicate energy differences of interest in materials chemistry and physics. The applications in this proposal are challenges for the simpler approximations of Kohn-Sham density functional theory, which are part of the current “standard model” for quantum chemistry and condensed matter physics. Direct RPA includes the full exact exchange energy and a nonlocal correlation energy from the occupied and unoccupied Kohn-Sham orbitals and orbital energies, with an approximate but universal description of long-range van der Waals attraction. RPA still cannot be a benchmark electronic structure method without a correction for its short-range correlation energy which is too deep. The short-range error of RPA can be corrected by a spatially nonlocal, exchange-only kernel. Such kernels could be a more reliable and “intrinsic” approximation to capture accurate ground-state energy differences than the expensive SOSEX and other beyond-RPA corrections. In this project we revealed the strengths and limitations of some model kernels through some materials problems focusing on structural phase transitions and transition metal chemistry. In addition we demonstrated the accuracy of the Renormalized RPA for total correlation energies of simple solids, and the pressure-induced phase transitions.

36 MATERIALS SCIENCE↗

Radiative transfer theory for passive microwave remote sensing of a two-layer random medium with cylindrical structures

A model of the vegetation layer as a two-layer random medium with a small correlation length l sub rho in the horizontal direction, and a large correlation length l sub z in the vertical direction, is presented for fields with cylindrical structures. As l sub z approaches infinity, closed form solutions are derived for the brightness temperatures; the kernels in the scattering terms of the radiative transfer equations result in delta functions showing that forward scattering is dominant over all other directions. The results are compared with the Gaussian quadrature method data for numerical solution of the radiative transfer equations.

Chuang, S. L.↗

VLSI Processor For Vector Quantization

Pixel intensities in each kernel compared simultaneously with all code vectors. Prototype high-performance, low-power, very-large-scale integrated (VLSI) circuit designed to perform compression of image data by vector-quantization method. Contains relatively simple analog computational cells operating on direct or buffered outputs of photodetectors grouped into blocks in imaging array, yielding vector-quantization code word for each such block in sequence. Scheme exploits parallel-processing nature of vector-quantization architecture, with consequent increase in speed.

Tawel, Raoul↗

Simulating and Detecting Radiation-Induced Errors for Onboard Machine Learning

Spacecraft processors and memory are subjected to high radiation doses and therefore employ radiation-hardened components. However, these components are orders of magnitude more expensive than typical desktop components, and they lag years behind in terms of speed and size. We have integrated algorithm-based fault tolerance (ABFT) methods into onboard data analysis algorithms to detect radiation-induced errors, which ultimately may permit the use of spacecraft memory that need not be fully hardened, reducing cost and increasing capability at the same time. We have also developed a lightweight software radiation simulator, BITFLIPS, that permits evaluation of error detection strategies in a controlled fashion, including the specification of the radiation rate and selective exposure of individual data structures. Using BITFLIPS, we evaluated our error detection methods when using a support vector machine to analyze data collected by the Mars Odyssey spacecraft. We found ABFT error detection for matrix multiplication is very successful, while error detection for Gaussian kernel computation still has room for improvement.

data analysis↗

Non-Uniform Contrast and Noise Correction for Coded Source Neutron Imaging

Since the first application of neutron radiography in the 1930s, the field of neutron radiography has matured enough to develop several applications. However, advances in the technology are far from concluded. In general, the resolution of scintillator-based detection systems is limited to the $10\mu m$ range, and the relatively low neutron count rate of neutron sources compared to other illumination sources restricts time resolved measurement. One path toward improved resolution is the use of magnification; however, to date neutron optics are inefficient, expensive, and difficult to develop. There is a clear demand for cost-effective scintillator-based neutron imaging systems that achieve resolutions of $1 \mu m$ or less. Such imaging system would dramatically extend the application of neutron imaging. For such purposes a coded source imaging system is under development. The current challenge is to reduce artifacts in the reconstructed coded source images. Artifacts are generated by non-uniform illumination of the source, gamma rays, dark current at the imaging sensor, and system noise from the reconstruction kernel. In this paper, we describe how to pre-process the coded signal to reduce noise and non-uniform illumination, and how to reconstruct the coded signal with three reconstruction methods correlation, maximum likelihood estimation, and algebraic reconstruction technique. We illustrates our results with experimental examples.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Comparative Study of Laboratory-Scale and Prototypic Production-Scale Fuel Fabrication Processes and Product Characteristics

An objective of the High Temperature Gas Reactor fuel development and qualification program for the United States Department of Energy has been to qualify fuel fabricated in prototypic production-scale equipment. The quality and characteristics of the tristructural isotropic coatings on fuel kernels are influenced by the equipment scale and processing parameters. Some characteristics affecting product quality were suppressed while others have become more significant in the larger equipment. Changes to the composition and method of producing resinated graphite matrix material has eliminated the use of hazardous, flammable liquids and enabled it to be procured as a vendor-supplied feed stock. A new method of overcoating TRISO particles with the resinated graphite matrix eliminates the use of hazardous, flammable liquids, produces highly spherical particles with a narrow size distribution, and attains product yields in excess of 99%. Compact fabrication processes have been scaled-up and automated with relatively minor changes to compact quality to manual laboratory-scale processes. The impact on statistical variability of the processes and the products as equipment was scaled are discussed. The prototypic production-scale processes produce test fuels that meet fuel quality specifications.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

A Data-Driven Global Sensitivity Analysis Framework for Three-Phase Distribution System with PVs

Global sensitivity analysis (GSA) of distribution systems with respect to stochastic PV and load variations plays an important role in designing optimal voltage control schemes. This paper proposes a data-driven framework for GSA of distribution systems. In particular, two representative surrogate modeling-based approaches are developed, including the traditional Gaussian process-based and the analysis of variance (ANOVA) kernel ones. The key idea is to develop a surrogate model that captures the hidden global relationship between voltage and real and reactive power injections from the historical data. With the surrogate model, the Sobol indices can be conveniently calculated through either the sampling-based method or the analytical method to assess the global sensitivity of voltage to variations of PV and load power injections. The sampling-based method approximates the Sobol indices using Monte Carlo simulations while the analytical method calculates them by resorting to the ANOVA expansion framework. Comparison results with other model-based GSA methods on the unbalanced three-phase IEEE 37-bus and 123-bus distribution systems show that the proposed framework can achieve much higher computational efficiency with negligible loss of accuracy. The results on a real 240-node distribution system using actual smart meter data further validate the feasibility and scalability of the proposed framework.

14 SOLAR ENERGY↗

A Formal Model for Real-Time Parallel Computation

The imposition of real-time constraints on a parallel computing environment--- specifically high-performance, cluster-computing systems--- introduces a variety of challenges with respect to the formal verification of the system's timing properties. In this paper, we briefly motivate the need for such a system, and we introduce an automaton-based method for performing such formal verification. We define the concept of a consistent parallel timing system: a hybrid system consisting of a set of timed automata (specifically, timed Buechi automata as well as a timed variant of standard finite automata), intended to model the timing properties of a well-behaved real-time parallel system. Finally, we give a brief case study to demonstrate the concepts in the paper: a parallel matrix multiplication kernel which operates within provable upper time bounds. We give the algorithm used, a corresponding consistent parallel timing system, and empirical results showing that the system operates under the specified timing constraints.

formal methods↗

Refinement of Methods for Evaluation of Near-Hypersingular Integrals in BEM Formulations

In this paper, we present advances in singularity cancellation techniques applied to integrals in BEM formulations that are nearly hypersingular. Significant advances have been made recently in singularity cancellation techniques applied to 1 R type kernels [M. Khayat, D. Wilton, IEEE Trans. Antennas and Prop., 53, pp. 3180-3190, 2005], as well as to the gradients of these kernels [P. Fink, D. Wilton, and M. Khayat, Proc. ICEAA, pp. 861-864, Torino, Italy, 2005] on curved subdomains. In these approaches, the source triangle is divided into three tangent subtriangles with a common vertex at the normal projection of the observation point onto the source element or the extended surface containing it. The geometry of a typical tangent subtriangle and its local rectangular coordinate system with origin at the projected observation point is shown in Fig. 1. Whereas singularity cancellation techniques for 1 R type kernels are now nearing maturity, the efficient handling of near-hypersingular kernels still needs attention. For example, in the gradient reference above, techniques are presented for computing the normal component of the gradient relative to the plane containing the tangent subtriangle. These techniques, summarized in the transformations in Table 1, are applied at the sub-triangle level and correspond particularly to the case in which the normal projection of the observation point lies within the boundary of the source element. They are found to be highly efficient as z approaches zero. Here, we extend the approach to cover two instances not previously addressed. First, we consider the case in which the normal projection of the observation point lies external to the source element. For such cases, we find that simple modifications to the transformations of Table 1 permit significant savings in computational cost. Second, we present techniques that permit accurate computation of the tangential components of the gradient; i.e., tangent to the plane containing the source element.

Fink, Patricia W.↗

SU-C-BRD-07: Three-Dimensional Dose Reconstruction in the Presence of Inhomogeneities Using Fast EPID-Based Back-Projection Method

Purpose: Three-dimensional dose verification can detect errors introduced by the treatment planning system (TPS) or differences between planned and delivered dose distribution during the treatment. The aim of the study is to extend a previous in-house developed three-dimensional dose reconstructed model in homogeneous phantom to situtions in which tissue inhomogeneities are present. Methods: The method was based on the portal grey images from an electronic portal imaging device (EPID) and the relationship between beamlets and grey-scoring voxels at the position of the EPID. The relationship was expressed in the form of grey response matrix that was quantified using thickness-dependence scatter kernels determined by series of experiments. From the portal grey-value distribution information measured by the EPID the two-dimensional incident fluence distribution was reconstructed based on the grey response matrix using a fast iterative algorithm. The accuracy of this approach was verified using a four-field intensity-modulated radiotherapy (IMRT) plan for the treatment of lung cancer in anthopomorphic phantom. Each field had between twenty and twenty-eight segments and was evaluated by comparing the reconstructed dose distribution with the measured dose. Results: The gamma-evaluation method was used with various evaluation criteria of dose difference and distance-to-agreement: 3%/3mm and 2%/2 mm. The dose comparison for all irradiated fields showed a pass rate of 100% with the criterion of 3%/3mm, and a pass rate of higher than 92% with the criterion of 2%/2mm. Conclusion: Our experimental results demonstrate that our method is capable of accurately reconstructing three-dimensional dose distribution in the presence of inhomogeneities. Using the method, the combined planning and treatment delivery process is verified, offing an easy-to-use tool for the verification of complex treatments.

60 APPLIED LIFE SCIENCES↗

On the eigenvector bias of Fourier feature networks: From regression to solving multi-scale PDEs with physics-informed neural networks

Physics-informed neural networks (PINNs) are demonstrating remarkable promise in integrating physical models with gappy and noisy observational data, but they still struggle in cases where the target functions to be approximated exhibit high-frequency or multi-scale features. Here in this work we investigate this limitation through the lens of Neural Tangent Kernel (NTK) theory and elucidate how PINNs are biased towards learning functions along the dominant eigen-directions of their limiting NTK. Using this observation, we construct novel architectures that employ spatio-temporal and multi-scale random Fourier features, and justify how such coordinate embedding layers can lead to robust and accurate PINN models. Numerical examples are presented for several challenging cases where conventional PINN models fail, including wave propagation and reaction–diffusion dynamics, illustrating how the proposed methods can be used to effectively tackle both forward and inverse problems involving partial differential equations with multi-scale behavior.

42 ENGINEERING↗

Effects of ray profile modeling on resolution recovery in clinical CT

Purpose: Iterative image reconstruction gains more and more interest in clinical routine, as it promises to reduce image noise (and thereby patient dose), to reduce artifacts, or to improve spatial resolution. However, among vendors and researchers, there is no consensus of how to best achieve these goals. The authors are focusing on the aspect of geometric ray profile modeling, which is realized by some algorithms, while others model the ray as a straight line. The authors incorporate ray-modeling (RM) in nonregularized iterative reconstruction. That means, instead of using one simple single needle beam to represent the x-ray, the authors evaluate the double integral of attenuation path length over the finite source distribution and the finite detector element size in the numerical forward projection. Our investigations aim at analyzing the resolution recovery (RR) effects of RM. Resolution recovery means that frequencies can be recovered beyond the resolution limit of the imaging system. In order to evaluate, whether clinical CT images can benefit from modeling the geometrical properties of each x-ray, the authors performed a 2D simulation study of a clinical CT fan-beam geometry that includes the precise modeling of these geometrical properties. Methods: All simulations and reconstructions are performed in native fan-beam geometry. A water phantom with resolution bar patterns and a Forbild thorax phantom with circular resolution patterns representing calcifications in the heart region are simulated. An FBP reconstruction with a Ram–Lak kernel is used as a reference reconstruction. The FBP is compared to iterative reconstruction techniques with and without RM: An ordered subsets convex (OSC) algorithm without any RM (OSC), an OSC where the forward projection is modeled concerning the finite focal spot and detector size (OSC-RM) and an OSC with RM and with a matched forward and backprojection pair (OSC-T-RM, T for transpose). In all cases, noise was matched to be able to focus on comparing spatial resolution. The authors use two different simulation settings. Both are based on the geometry of a typical clinical CT system (0.7 mm detector element size at isocenter, 1024 projections per rotation). Setting one has an exaggerated source width of 5.0 mm. Setting two has a realistically small source width of 0.5 mm. The authors also investigate the transition from setting one to two. To quantify image quality, the authors analyze line profiles through the resolution patterns to define a contrast factor (CF) for contrast-resolution plots, and the authors compare the normalized cross-correlation (NCC) with respect to the ground truth of the circular resolution patterns. To independently analyze whether RM is of advantage, the authors implemented several iterative reconstruction algorithms: The statistical iterative reconstruction algorithm OSC, the ordered subsets simultaneous algebraic reconstruction technique (OSSART) and another statistical iterative reconstruction algorithm, denoted with ordered subsets maximum likelihood (OSML) algorithm. All algorithms were implemented both without RM (denoted as OSC, OSSART, and OSML) and with RM (denoted as OSC-RM, OSSART-RM, and OSML-RM). Results: For the unrealistic case of a 5.0 mm focal spot the CF can be improved by a factor of two due to RM: the 4.2 LP/cm bar pattern, which is the first bar pattern that cannot be resolved without RM, can be easily resolved with RM. For the realistic case of a 0.5 mm focus, all results show approximately the same CF. The NCC shows no significant dependency on RM when the source width is smaller than 2.0 mm (as in clinical CT). From 2.0 mm to 5.0 mm focal spot size increasing improvements can be observed with RM. Conclusions: Geometric RM in iterative reconstruction helps improving spatial resolution, if the ray cross-section is significantly larger than the ray sampling distance. In clinical CT, however, the ray is not much thicker than the distance between neighboring ray centers, as the focal spot size is small and detector crosstalk is negligible, due to reflective coatings between detector elements. Therefore,RM appears not to be necessary in clinical CT to achieve resolution recovery.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

A comprehensive modeling procedure for estimating statistical properties of forced ignition

In this work, a comprehensive modeling procedure for estimating the probability of ignition with application to high altitude relights of aircraft combustors is developed. In these configurations, an ignitor is used to introduce high-enthalpy discharge into a fuel-laden but stratified flow. Because of the inherent variabilities inflow conditions, kernel discharge process, and the chaotic turbulent flow, ignition cannot be described deterministically, but only as a probabilistic measure. The proposed modeling framework consists of three components. The turbulent flow is represented using the large eddy simulation (LES) framework. The ignition process is modeled using a manifold approach, where the initial kernel evolution is represented using a homogeneous reactor while the latter part of the evolution is represented as a competition between diffusion and chemical reactions using a flamelet-type mapping. A combined lookup table that can track the evolution of the kernel through these two distinct reaction stages is developed. The table lookup variables are solved within the LES framework. The resulting simulation tool is then embedded within an uncertainty quantification approach, where variations in the turbulent flow, as well as operating and kernel properties, are simulated using a Monte-Carlo-based sampling approach. Techniques to reduce computational cost are used to obtain a robust, numerically accurate, and physically representative model for engine relight. The method is validated using experimental data for ignition of methane/air mixtures. Due to the comprehensive nature of the modeling procedure, it is found that the simulation tool reproduces experimentally observed ignition probabilities over a wide range of operating conditions.

42 ENGINEERING↗

A graphics processing unit accelerated sparse direct solver and preconditioner with block low rank compression

We present the GPU implementation efforts and challenges of the sparse solver package STRUMPACK. The code is made publicly available on github with a permissive BSD license. STRUMPACK implements an approximate multifrontal solver, a sparse LU factorization which makes use of compression methods to accelerate time to solution and reduce memory usage. Multiple compression schemes based on rank-structured and hierarchical matrix approximations are supported, including hierarchically semi-separable, hierarchically off-diagonal butterfly, and block low rank. Here, in this paper, we present the GPU implementation of the block low rank (BLR) compression method within a multifrontal solver. Our GPU implementation relies on highly optimized vendor libraries such as cuBLAS and cuSOLVER for NVIDIA GPUs, rocBLAS and rocSOLVER for AMD GPUs and the Intel oneAPI Math Kernel Library (oneMKL) for Intel GPUs. Additionally, we rely on external open source libraries such as SLATE (Software for Linear Algebra Targeting Exascale), MAGMA (Matrix Algebra on GPU and Multi-core Architectures), and KBLAS (KAUST BLAS). SLATE is used as a GPU-capable ScaLAPACK replacement. From MAGMA we use variable sized batched dense linear algebra operations such as GEMM, TRSM and LU with partial pivoting. KBLAS provides efficient (batched) low rank matrix compression for NVIDIA GPUs using an adaptive randomized sampling scheme. The resulting sparse solver and preconditioner runs on NVIDIA, AMD and Intel GPUs. Interfaces are available from PETSc, Trilinos and MFEM, or the solver can be used directly in user code. We report results for a range of benchmark applications, using the Perlmutter system from NERSC, Frontier from ORNL, and Aurora from ALCF. For a high frequency wave equation on a regular mesh, using 32 Perlmutter compute nodes, the factorization phase of the exact GPU solver is about 6.5× faster compared to the CPU-only solver. The BLR-enabled GPU solver is about 13.8× faster than the CPU exact solver. For a collection of SuiteSparse matrices, the STRUMPACK exact factorization on a single GPU is on average 1.9× faster than NVIDIA’s cuDSS solver.

97 MATHEMATICS AND COMPUTING↗

Userspace Squash Filesystem for Launching Linux Containers on HPC Systems [Thesis]

The demand for user defined software stacks (UDSS) has been increasing in the high-performance computing (HPC) community. Container technology has become popular due to the flexibility and isolation it provides to HPC users. Container images must be available to all nodes involved for use in HPC and can be distributed to compute nodes in a variety of ways. A common method for container image distribution is to simply copy the container image to memory on each compute node, which can be time consuming at scale, and uses valuable memory on each node. The kernel mounted squash filesystem (squashfs) has proven fast and efficient for this task but requires root-level access. This paper will show a user space mounted squashfs is an efficient and secure solution for container image distribution.

97 MATHEMATICS AND COMPUTING↗

Lidar conversion parameters derived from SAGE II extinction measurements

SAGE II multiwavelength aerosol extinction measurements are used to estimate mass- and extinction-to-backscatter conversion parameters. The basis of the analysis is the principal component analysis of the SAGE II extinction kernels to estimate both total aerosol mass and aerosol backscatter at a variety of wavelengths. Comparisons of coincident SAGE II extinction profiles with 0.694-micron aerosol backscatter profiles demonstrate the validity of the method.

Thomason, L. W.↗