Search NASA⌕ Search

SEARCH · Search NASA

Results for “eigenvalues”

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 649 records · Page 36

Fresh-Core Reload of the Neutron Radiography (NRAD) Reactor with Uranium(20)-Erbium-Zirconium-Hydride Fuel

The neutron radiography (NRAD) reactor is a 250 kW TRIGA® (Training, Research, Isotopes, General Atomics) Mark II , tank-type research reactor currently located in the basement, below the main hot cell, of the Hot Fuel Examination Facility (HFEF) at the Idaho National Laboratory (INL). It is equipped with two beam tubes with separate radiography stations for the performance of neutron radiography irradiation on small test components. The 60-fuel-element operational core configuration of the NRAD LEU TRIGA reactor has been evaluated as an acceptable benchmark experiment. The initial critical configuration developed during the fuel loading process, which contains only 56 fuel elements, has not been evaluated as it is very similar to the evaluated core configuration. The benchmark eigenvalue is 1.0012 ± 0.0029. Calculated eigenvalues differ significantly (~±1%) from the benchmark eigenvalue and have demonstrated sensitivity to the thermal scattering treatment of hydrogen in the U-Er-Zr-H fuel.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

MPACT Subgroup Self-Shielding Efficiency Improvements

Recent developments to improve the efficiency of the MOC solvers in MPACT have yielded effective kernels that loop over several energy groups at once, rather that looping over one group at a time. These kernels have produced roughly a 2x speedup on the MOC sweeping time during eigenvalue calculation. However, the self-shielding subgroup calculation had not been reevaluated to take advantage of these new kernels, which typically requires substantial solve time. The improvements covered in this report start by integrating the multigroup kernel concepts into the subgroup calculation, which are then used as the basis for further extensions. The next improvement that is covered is what is currently being termed as “Lumped Parameter MOC”. Because the subgroup calculation is a purely fixed source problem and multiple sweeps are performed only to update the boundary angular fluxes, the sweep procedure can be condensed to allow for the instantaneous propagation of the flux across a spatial domain, without the need to sweep along all segments in a ray. Once the boundary angular fluxes are considered to be converged, an additional sweep that will tally the scalar flux is completed. The last improvement that is investigated is the possible reduction of the number of azimuthal angles per octant in the shielding sweep. Typically 16 azimuthal angles per octant are used for self-shielding and eigenvalue calculations, but it is possible that the self-shielding sweeps are less sensitive to the number of angles than the full eigenvalue calculation.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

The Cosmic V-Web

The network of filaments with embedded clusters surrounding voids, which has been seen in maps derived from redshift surveys and reproduced in simulations, has been referred to as the cosmic web. A complementary description is provided by considering the shear in the velocity field of galaxies. The eigenvalues of the shear provide information regarding whether or not a region is collapsing in three dimensions, which is the condition for a knot, expanding in three dimensions, which is the condition for a void, or in the intermediate condition of a filament or sheet. The structures that are quantitatively defined by the eigenvalues can be approximated by iso-contours that provide a visual representation of the cosmic velocity (V) web. The current application is based on radial peculiar velocities from the Cosmicflows-2 collection of distances. The three-dimensional velocity field is constructed using the Wiener filter methodology in the linear approximation. Eigenvalues of the velocity shear are calculated at each point on a grid. Here, knots and filaments are visualized across a local domain of diameter ∼0.1c.

79 ASTRONOMY AND ASTROPHYSICS↗

Extending the eigCG algorithm to non-symmetric linear systems with multiple right-hand sides

For Hermitian positive definite linear systems and eigenvalue problems, the eigCG algorithm is a memory efficient algorithm that solves the linear system and simultaneously computes some of its eigenvalues. The algorithm is based on the Conjugate-Gradient (CG) algorithm, however, it uses only a window of the vectors generated by the CG algorithm to compute approximate eigenvalues. The number and accuracy of the eigenvectors can be increased by solving more right-hand sides. For Hermitian systems with multiple right-hand sides, the computed eigenvectors can be used to speed up the solution of subsequent systems. The algorithm was tested on Lattice QCD problems by solving the normal equations and was shown to give large speed up factors and to remove the critical slowing down as we approach light quark masses. Here, an extension to the non-symmetric case based on the two-sided Lanczos algorithm is given. The new algorithm is tested on Lattice QCD problems and is shown to give promising results. We also study the removal of the critical slowing down and compare results with those of the eigCG algorithm. We also discuss the case when the system is gamma5-Hermitian.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Whole-core comet solutions to a 3-dimensional PWR benchmark problem with gadolinium

A pressurized water reactor (PWR) benchmark problem with gadolinium was used to determine the accuracy and computational efficiency of the coarse mesh radiation transport method COMET. The benchmark problem contains 193 square fuel assemblies. The COMET solution (eigenvalue, assembly averaged and fuel pin averaged fission density distributions) was compared with those obtained from the corresponding Monte Carlo reference solution using the same 2-group material cross section library. The comparison showed that both the core eigenvalue and fission density distribution averaged over each assembly and fuel pin predicated by COMET agree very well with the corresponding MCNP reference solution if the incident flux response expansion used in COMET is truncated at 2nd order in the two spatial and the two angular variables. The benchmark calculations indicate that COMET has Monte Carlo accuracy. In, particular, the eigenvalue difference between the codes ranged from 17 pcm to 35 pcm, being within 2 standard deviations of the calculational uncertainty. The mean flux weighted relative differences in the assembly and fuel pin fission densities were 0.47% and 0.65%, respectively. It was also found that COMET's full (whole) core computational speed is 30,000 times faster than MCNP in which only 1/8 of the core is modeled. It is estimated that COMET would have been about over 6 orders of magnitude faster than MCNP if the full core were also modeled in MCNP. (authors)

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Verification of the shift Monte Carlo code with the C5G7 reactor benchmark

Shift is a new hybrid Monte Carlo/deterministic radiation transport code being developed at Oak Ridge National Laboratory. At its current stage of development, Shift includes a parallel Monte Carlo capability for simulating eigenvalue and fixed-source multigroup transport problems. This paper focuses on recent efforts to verify Shift's Monte Carlo component using the two-dimensional and three-dimensional C5G7 NEA benchmark problems. Comparisons were made between the benchmark eigenvalues and those output by the Shift code. In addition, mesh-based scalar flux tally results generated by Shift were compared to those obtained using MCNP5 on an identical model and tally grid. The Shift-generated eigenvalues were within three standard deviations of the benchmark and MCNP5-1.60 values in all cases. The flux tallies generated by Shift were found to be in very good agreement with those from MCNP. (authors)

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Optimization of multi-group cross sections for fast reactor analysis

The selection of the number of broad energy groups, collapsed broad energy group boundaries, and their associated evaluation into collapsed macroscopic cross sections from a general 238-group ENDF/B-VII library dramatically impacted the k eigenvalue for fast reactor analysis. An analysis was undertaken to assess the minimum number of energy groups that would preserve problem physics; this involved studies using the 3D deterministic transport parallel code PENTRAN, the 2D deterministic transport code SCALE6.1, the Monte Carlo based MCNP5 code, and the YGROUP cross section collapsing tool on a spatially discretized MOX fuel pin comprised of 21% PUO{sub 2}-UO{sub 2} with sodium coolant. The various cases resulted in a few hundred pcm difference between cross section libraries that included the 238 multi-group reference, and cross sections rendered using various reaction and adjoint weighted cross sections rendered by the YGROUP tool, and a reference continuous energy MCNP case. Particular emphasis was placed on the higher energies characteristic of fission neutrons in a fast spectrum; adjoint computations were performed to determine the average per-group adjoint fission importance for the MOX fuel pin. This study concluded that at least 10 energy groups for neutron transport calculations are required to accurately predict the eigenvalue for a fast reactor system to within 250 pcm of the 238 group case. In addition, the cross section collapsing/weighting schemes within YGROUP that provided a collapsed library rendering eigenvalues closest to the reference were the contribution collapsed, reaction rate weighted scheme. A brief analysis on homogenization of the MOX fuel pin is also provided, although more work is in progress in this area. (authors)

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

A comparison between the fission matrix method, the diffusion model and the transport model

The fission matrix method may be used to solve the critical eigenvalue problem in a Monte Carlo simulation. This method gives us access to the different eigenvalues and eigenvectors of the transport or fission operator. We propose to compare the results obtained via the fission matrix method with those of the diffusion model, and an approximated transport model. To do so, we choose to analyse the mono-kinetic and continuous energy cases for a Godiva-inspired critical sphere. The first five eigenvalues are computed with TRIPOLI-4{sup R} and compared to the theoretical ones. An extension of the notion of the extrapolation distance is proposed for the modes other than the fundamental one. (authors)

97 MATHEMATICS AND COMPUTING↗

Verification of the pin power reconstruction method in NESTLE

NESTLE is a 3D nodal core simulator that solves the few group neutron diffusion equation using the nodal expansion method (NEM) for the steady state (eigenvalue or fixed-source), adjoint (eigenvalue), and transient (eigenvalue or fixed-source) problems. NESTLE has been in use for over 25 years and has been continuously improved and developed. NESTLE can be used to model either Cartesian geometries (for use with typical PWR, BWR fuel assemblies) or hexagonal geometries (for use with VVER and other triangularly pitched fuel assemblies). As part of NESTLE's ongoing verification and validation process, several computational comparisons have been made to quantify the discrepancies in NESTLE's hexagonal pin power reconstruction method. In this paper, we present the results of these code-to-code comparisons and find that the root mean squared (RMS) error in pin power is comparable to other nodal codes.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Applying Nonlinear Diffusion Acceleration to Fixed-Source Problems with Anisotropic Scattering - Paper 15

Work over the last several years has demonstrated the impressive performance gains that can be attained by using Moment-Based Acceleration (MBA), or High-Order/Low-Order (HOLO), methods in a variety of transport applications. In the field of neutronics, Nonlinear Diffusion Acceleration (NDA) has been used to accelerate the solution to the fixed-source problem and the k-eigenvalue problem. These methods efficiently move a major portion of the computational work out of the full phase-space (high-order (HO)) domain into a reduced phase-space (low-order (LO)) setting. In previous implementations, NDA has only been applied to isotropic scattering problems. In this paper, we will extend this methodology to anisotropic scattering. This initial work on anisotropic scattering will focus on the 1-D multi-group neutron transport equation. This summary focuses on a fixed-source problem, but the extension to criticality problems is straight-forward via NDA-NCA. Within this paper, we have described a method for extending Nonlinear Diffusion Acceleration to solve problems with anisotropic scattering. We demonstrated that a consistent low-order system, resembling the low-order system from, could be constructed. We demonstrated that the NDA method performed exceptionally in the presence of anisotropic scattering. NDA reduced the number of iterations required to meet the convergence tolerance by a factor of 500 over source iteration for a one fixed-source problem and by 3.5-50 x for the Henyey-Greenstein problems. This method is equally well-suited for eigenvalue problems. The NDA method allows us to transition the eigenvalue iteration to the low-order domain.

61 RADIATION PROTECTION AND DOSIMETRY↗

MODIFIED CHOKE FLOW CRITERION FOR THE TWO-PHASE TWO-FLUID MODEL

A choked condition exists when mass flow rate becomes independent of the downstream conditions. In other words, no information can propagate in the upstream direction under this condition. The real part of the solution of the characteristic equation for the model represents velocity of the signal propagation and the imaginary part is the growth (or decay) rate of that signal. Therefore, if the real part of these eigenvalues is positive then no signal propagates in the upstream direction (choosing downstream direction to be the positive direction) resulting in the choke flow. In order to develop the choke criterion, a non-dimensional form of the characteristic equation is derived for the standard two-phase two-fluid model. The equation is in the terms of a slip Mach number Ms. It can be shown that the slip Mach number is small for many applications including nuclear reactor safety simulations. The eigenvalues of the characteristic equation are obtained as a power series expansion about the point Ms = 0. These eigenvalues are used to develop a choking criterion for the compressible two-phase flows.

97 MATHEMATICS AND COMPUTING↗

Numerical solution of singular Lyapunov equations

We consider the numerical solution of large scale singular (continuous-time) Lyapunov equations of the form AX + XA T + BB T = 0, where A is semistable, that is, its spectrum is contained in the left half plane, with the exception of a few semisimple eigenvalues at zero. We also consider the case of a few semisimple eigenvalues on the imaginary axis. We assume that we know these few eigenvalues (zero or imaginary), and that we have or can compute the corresponding invariant subspaces. We use this information to build an appropriate newly proposed subspace on which to project the Lyapunov equations, and then compute a low-rank approximation to the least squares solution. Selected illustrative numerical examples are provided.

97 MATHEMATICS AND COMPUTING↗

Machine Learning Moment Closure Models for the Radiative Transfer Equation III: Enforcing Hyperbolicity and Physical Characteristic Speeds

This is the third paper in a series in which we develop machine learning (ML) moment closure models for the radiative transfer equation. In our previous work (Huang et al. in J Comput Phys 453:110941, 2022), we proposed an approach to learn the gradient of the unclosed high order moment, which performs much better than learning the moment itself and the conventional $P_N$ closure. However, while the ML moment closure has better accuracy, it is not able to guarantee hyperbolicity and has issues with long time stability. In our second paper (Huang et al., in: Machine learning moment closure models for the radiative transfer equation II: enforcing global hyperbolicity in gradient based closures, 2021. arXiv:2105.14410), we identified a symmetrizer which leads to conditions that enforce that the gradient based ML closure is symmetrizable hyperbolic and stable over long time. The limitation of this approach is that in practice the highest moment can only be related to four, or fewer, lower moments. In this paper, we propose a new method to enforce the hyperbolicity of the ML closure model. Motivated by the observation that the coefficient matrix of the closure system is a lower Hessenberg matrix, we relate its eigenvalues to the roots of an associated polynomial. Here, we design two new neural network architectures based on this relation. The ML closure model resulting from the first neural network is weakly hyperbolic and guarantees the physical characteristic speeds, i.e., the eigenvalues are bounded by the speed of light. The second model is strictly hyperbolic and does not guarantee the boundedness of the eigenvalues. Several benchmark tests including the Gaussian source problem and the two-material problem show the good accuracy, stability and generalizability of our hyperbolic ML closure model.

97 MATHEMATICS AND COMPUTING↗

Orbital-enriched flat-top partition of unity method for the Schrödinger eigenproblem

Quantum mechanical calculations require the repeated solution of a Schrödinger equation for the wavefunctions of the system, from which materials properties follow. Recent work has shown the effectiveness of enriched finite element type Galerkin methods at significantly reducing the degrees of freedom required to obtain accurate solutions. However, time to solution has been adversely affected by the need to solve a generalized rather than standard eigenvalue problem and the ill-conditioning of associated system matrices. In this work, we address both issues by proposing a stable and efficient orbital-enriched partition of unity method to solve the Schrödinger boundary-value problem in a parallelepiped unit cell subject to Bloch-periodic boundary conditions. In the proposed partition of unity method, the three-dimensional domain is covered by overlapping patches, with a compactly-supported weight function associated with each patch. A key ingredient in our approach is the use of non-negative weight functions that possess the flat-top property, i.e., each weight function is identically equal to unity over some finite subset of its support. This flat-top property provides a pathway to devise a stable approximation over the whole domain. On each patch, we use $p$th degree orthogonal (Legendre) polynomials that ensure $p$th order completeness, and in addition include eigenfunctions of the radial Schrödinger equation. Furthermore, we adopt a variational lumping approach to construct a (block-)diagonal overlap matrix that yields a standard eigenvalue problem for which there exist efficient eigensolvers. The accuracy, stability, and efficiency of the proposed method is demonstrated for the Schrödinger equation with a harmonic potential as well as a localized Gaussian potential. In conclusion, we show that the proposed approach delivers optimal rates of convergence in the energy, and the use of orbital enrichment significantly reduces the number of degrees of freedom for a given desired accuracy in the energy eigenvalues while the stability of the enriched approach is fully maintained.

42 ENGINEERING↗

Rethinking the ill-posedness of the spectral function reconstruction — Why is it fundamentally hard and how Artificial Neural Networks can help

Reconstructing hadron spectral functions through Euclidean correlation functions are of the important missions in lattice QCD calculations. However, in a Källen–Lehmann (KL) spectral representation, the reconstruction is observed to be ill-posed in practice. It is usually ascribed to the fewer observation points compared to the number of points in the spectral function. In this paper, by solving the eigenvalue problem of continuous KL convolution, we show analytically that the ill-posedness of the inversion is fundamental and it exists even for continuous correlation functions. We discussed how to introduce regulators to alleviate the predicament, in which include the Artificial Neural Networks (ANNs) representations recently proposed by the Authors in another study. The uniqueness of solutions using ANNs representations is manifested analytically and validated numerically. Reconstructed spectral functions using different regularization schemes are also demonstrated, together with their eigen-mode decomposition. We observe that components with large eigenvalues can be reliably reconstructed by all methods, whereas those with low eigenvalues need to be constrained by regulators.

97 MATHEMATICS AND COMPUTING↗

Novel strategies for modal-based structural material identification

Here, we present modal-based methods for model calibration in structural dynamics, and address several key challenges in the solution of gradient-based optimization problems with eigenvalues and eigenvectors, including the solution of singular Helmholtz problems encountered in sensitivity calculations, non-differentiable objective functions caused by mode swapping during optimization, and cases with repeated eigenvalues. Unlike previous literature that relied on direct solution of the eigenvector adjoint equations, we present a parallel iterative domain decomposition strategy (Adjoint Computation via Modal Superposition with Truncation Augmentation) for the solution of the singular Helmholtz problems. For problems with repeated eigenvalues we present a novel Mode Separation via Projection algorithm, and in order to address mode swapping between inverse iterations we present a novel Injective mode ordering metric. We present the implementation of these methods in a massively parallel finite element framework with the ability to use measured modal data to extract unknown structural model parameters from large complex problems. A series of increasingly complex numerical examples are presented that demonstrate the implementation and performance of the methods in a massively parallel finite element framework [7], [5], using gradient-based optimization techniques in the Rapid Optimization Library (ROL) [21].

36 MATERIALS SCIENCE↗

Model Order Reduction Algorithm for Estimating the Absorption Spectrum

The ab initio description of the spectral interior of the absorption spectrum poses both a theoretical and computational challenge for modern electronic structure theory. Due to the often spectrally dense character of this domain in the quantum propagator’s eigenspectrum for medium-to-large sized systems, traditional approaches based on the partial diagonalization of the propagator often encounter oscillatory and stagnating convergence. Electronic structure methods which solve the molecular response problem through the solution of spectrally shifted linear systems, such as the complex polarization propagator, offer an alternative approach which is agnostic to the underlying spectral density or domain location. This generality comes at a seemingly high computational cost associated with solving a large linear system for each spectral shift in some discretization of the spectral domain of interest. In this work, we present a novel, adaptive solution to this high computational overhead based on model order reduction techniques via interpolation. Model order reduction reduces the computational complexity of mathematical models and is ubiquitous in the simulation of dynamical systems and control theory. The efficiency and effectiveness of the proposed algorithm in the ab initio prediction of X-ray absorption spectra is demonstrated using a test set of challenging water clusters which are spectrally dense in the neighborhood of the oxygen K-edge. On the basis of a single, user defined tolerance we automatically determine the order of the reduced models and approximate the absorption spectrum up to the given tolerance. We also illustrate that, for the systems studied, the automatically determined model order increases logarithmically with the problem dimension, compared to a linear increase of the number of eigenvalues within the energy window. Furthermore, we observed that the computational cost of the proposed algorithm only scales quadratically with respect to the problem dimension.

IDREAM, K-Edge, theoretical, computational, electr↗

Computational Aspects of Single-Molecule Kinetics for Coupled Catalytic Cycles: A Spectral Analysis

Catalysis from single active sites is analyzed using methods developed from single molecule kinetics. Using a stochastic Markov state description, the observable properties of general catalytic networks of reactions are expressed using an eigenvalue decomposition of the transition matrix for the Markov process. By the use of a sensitivity analysis, the necessary eigenvalues and eigenvectors are related to the energies of controlling barriers and wells located along the reaction routes. A generalization of the energetic span theory allows the eigenvalues to be computed from several activation energies corresponding to distinct barrier-well pairings. The formalism is demonstrated accurately for model problems for a physically realistic mechanism for an alkene hydrogenation reaction on a single atom catalyst.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗