Search NASA⌕ Search

SEARCH · Search NASA

Results for “finite difference 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 991 records · Page 55

Erratum: “A finite difference Davidson procedure to sidestep full ab initio hessian calculation: Application to characterization of stationary points and transition state searches”

Certain key contributions to the field were not presented in the Introduction section of our recent paper. In addition to the variational method described in Ref. 14 and iterative diagonalization methods in Refs. 15–17, we wish to direct the attention of the reader to other significant efforts that have utilized similar principles for various applications. Reiher and Neugebauer developed a mode-tracking algorithm using the Davidson method for calculating only the modes relevant for vibrational analysis. Similarly, a block-Davidson approach via finite differentiation of the gradient was implemented for calculating hessian-vector products for normal mode analysis. Deglmann and Furche developed a method for stationary point characterization with density functional theory where full solution of all coupled-perturbed Kohn-Sham equations, the bottleneck in hessian calculation, could be avoided through iterative calculation of hessian-vector products to determine only the lowest eigenvalues.

99 GENERAL AND MISCELLANEOUS↗

Customizable two-species kinetic equilibria for nonuniform low-beta plasmas

Here, two methods are developed for constructing self-consistent two-species kinetic equilibria for low-beta electrostatic plasmas, in which the magnetic field is uniform and fixed in time. The first method uses Taylor series approximations to construct distribution functions that can be specified analytically. The second method relies on numerically solving a nonlinear ordinary differential equation and produces exact—to numerical precision—equilibria. In both methods, the equilibrium distribution functions for ions and electrons are expressed in terms of constants of motion and satisfy the steady-state Vlasov-Poisson equation system. Provided that the ion drift speed does not exceed the ion thermal speed, the equilibria can be specified with customizable density and electrostatic potential profiles. The methods can thereby be tailored to different applications and are successfully applied to construct kinetic equilibria for cross-field plasmas with sheared flows, large density variations, and different levels of magnetization. The equilibria are used to initialize fourth-order finite-volume Vlasov-Poisson simulations in (x, v x , v y ) coordinates and the associated temporal evolution is used to assess the accuracy of each method. Lastly, the low-amplitude deviations observed in these simulations demonstrate that the kinetic equilibria are robust and that they provide a valuable means of studying the dynamics of nonuniform magnetized plasmas.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Heat and Mass Transfer in Cesium Evaporation from Graphite Surface in Argon

The aim of the present work is mathematical modeling of the heat and mass transfer of {sup 137}Cs under the action of low-temperature equilibrium plasma flow on sections of graphite contaminated by radionuclides. The methods of investigation are calculation of the temperature and mass rate of evaporation of radiocesium by solving a system of quasi-homogeneous nonstationary heat conduction equations with boundary conditions of the first and third kind, which take account of mass transfer, by means of finite-difference elements using the ANSYS package.

42 ENGINEERING↗

M-Adapting Low Order Mimetic Finite Differences for Dielectric Interface Problems

We consider a problem of reducing numerical dispersion for electromagnetic wave in the domain with two materials separated by a at interface in 2D with a factor of two di erence in wave speed. The computational mesh in the homogeneous parts of the domain away from the interface consists of square elements. Here the method construction is based on m-adaptation construction in homogeneous domain that leads to fourth-order numerical dispersion (vs. second order in non-optimized method). The size of the elements in two domains also di ers by a factor of two, so as to preserve the same value of Courant number in each. Near the interface where two meshes merge the mesh with larger elements consists of degenerate pentagons. We demonstrate that prior to m-adaptation the accuracy of the method falls from second to rst due to breaking of symmetry in the mesh. Next we develop m-adaptation framework for the interface region and devise an optimization criteria. We prove that for the interface problem m-adaptation cannot produce increase in method accuracy. This is in contrast to homogeneous medium where m-adaptation can increase accuracy by two orders.

97 MATHEMATICS AND COMPUTING↗

Phase retarder based on one-dimensional photonic crystals composed of plasma and mu-negative materials

By using transfer matrix method, a systematic study on the properties of the reflection phase shifts and the reflection phase difference between TE and TM waves in a finite one-dimensional (1D) photonic crystal containing plasma and mu-negative materials is presented. It is found that the reflection phase difference between the two polarizations remains constant in a rather wide frequency range for a given incident angle. More specifically, the reflection phase difference increases gradually from 0 to π rad with the increase of the incident angle. That is to say, the finite 1D structure can serve as a broadband phase retarder. It is also evident that the working frequency range of the phase retarder can be adjusted by altering the plasma frequency and the thickness of the plasma layers without changing the structure of the photonic crystal.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

A Characteristics Approach to the Finite Element Method

Herein, we present a new method for solving the linear Boltzmann transport equation. Two commonly used and well-understood methods for solving partial differential equations are the method of characteristics (MOC) and the finite element method (FEM). We propose a new method that combines the fundamental concept of the FEM with the analytic solution from the MOC to obtain coefficients for the FEM basis function expansion. Traditionally, coefficients for the FEM basis function expansion are obtained via matrix inversion. Instead, we solve for the coefficients with the MOC and represent the underlying fields with the basis function expansion using these coefficients. We provide a convergence study for our method with results from two sets of FEM basis functions: Gauss-Legendre and Gauss-Lobatto sets. We also compare two different variations of our method categorized as short characteristics and intermediate characteristics.

42 ENGINEERING↗

FELIX-2.0: New version of the finite element solver for the time dependent generator coordinate method with the Gaussian overlap approximation

The time-dependent generator coordinate method (TDGCM) is a powerful method to study the largeamplitude collective motion of quantum many-body systems such as atomic nuclei. Under the GaussianOverlap Approximation (GOA), the TDGCM leads to a local, time-dependent Schrödinger equation in amulti-dimensional collective space. In this paper, we present the version 2.0 of the code FELIX that solvesthe collective Schrödinger equation in a finite element basis. This new version features: (i) the ability tosolve a generalized TDGCM+GOA equation with a metric term in the collective Hamiltonian, (ii) supportfor new kinds of finite elements and different types of quadrature to compute the discretized Hamiltonianand overlap matrices, (iii) the possibility to leverage the spectral element scheme, (iv) an explicit Krylovapproximation of the time propagator for time integration instead of the implicit Crank–Nicolson methodimplemented in the first version, (v) an entirely redesigned workflow. We benchmark this release on ananalytic problem as well as on realistic two-dimensional calculations of the low-energy fission of240Puand256Fm. Low to moderate numerical precision calculations are most efficiently performed with simplexelements with a degree 2 polynomial basis. Higher precision calculations should instead use the spectralelement method with a degree 4 polynomial basis. We emphasize that in a realistic calculation of fissionmass distributions of240Pu, FELIX-2.0 is about 20 times faster than its previous release (within a numericalprecision of a few percents).

Regnier, D.↗

Optimal Control of a Degenerate PDE for Surface Shape

Controlling the shapes of surfaces provides a novel way to direct self-assembly of colloidal particles on those surfaces and may be useful for material design. This motivates the investigation of an optimal control problem for surface shape in this paper. Specifically, we consider an objective (tracking) functional for surface shape with the prescribed mean curvature equation in graph form as a state constraint. The control variable is the prescribed curvature. We prove existence of an optimal control, and using improved regularity estimates, we show sufficient differentiability to make sense of the first order optimality conditions. This allows us to rigorously compute the gradient of the objective functional for both the continuous and discrete (finite element) formulations of the problem. Numerical results are shown to illustrate the minimizers and optimal controls on different domains.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Nonlocal symmetries of the Hirota-Satsuma coupled Korteweg-de Vries system and their applications: Exact interaction solutions and integrable hierarchy

The nonlocal symmetry is derived from the known Darboux transformation (DT) of the Hirota-Satsuma coupled Korteweg-de Vries (HS-cKdV) system, and infinitely many nonlocal symmetries are given by introducing the internal parameters. By extending the HS-cKdV system to an auxiliary system with five dependent variables, the prolongation is found to localize the so-called seed nonlocal symmetry related to the DT. By applying the general Lie point symmetry method to this enlarged system, we obtain two main results: a new type of finite symmetry transformation is derived, which is different from the initial DT and can generate new solutions from old ones; some novel exact interaction solutions among solitons and other complicated waves including periodic cnoidal waves and Painlevé waves are computed through similarity reductions. In addition, two kinds of new integrable models are proposed from the obtained nonlocal symmetry: the negative HS-cKdV hierarchy by introducing the internal parameters; the integrable models both in lower and higher dimensions by restricting the symmetry constraints.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Validation of Finite-Element Models of Persistent-Current Effects in Nb 3 Sn Accelerator Magnets

Persistent magnetization currents are induced in superconducting filaments during the current ramping in magnets. The resulting perturbation to the design magnetic field leads to field quality degradation, in particular at low field where the effect is stronger relative to the main field. The effects observed in NbTi accelerator magnets were reproduced well with the critical-state model. However, this approach becomes less accurate for the calculation of the persistent-current effects observed in Nb 3 Sn accelerator magnets. Here a finite-element method based on the measured strand magnetization is validated against three state-of-art Nb3Sn accelerator magnets featuring different subelement diameters, critical currents, magnet designs and measurement temperatures. The temperature dependence of the persistent-current effects is reproduced. Based on the validated model, the impact of conductor design on the persistent current effects is discussed. The performance, limitations and possible improvements of the approach are also discussed.

36 MATERIALS SCIENCE↗

Ultrasound viscoelasticity assessment using an adaptive torsional shear wave propagation method

Purpose: Different approaches have been used in dynamic elastography to assess mechanical properties of biological tissues. Most techniques are based on a simple inversion based on the measurement of the shear wave speed to assess elasticity, whereas some recent strategies use more elaborated analytical or finite element method (FEM) models. In this study, a new method is proposed for the quantification of both shear storage and loss moduli of confined lesions, in the context of breast imaging, using adaptive torsional shear waves (ATSWs) generated remotely with radiation pressure. Methods: A FEM model was developed to solve the inverse wave propagation problem and obtain viscoelastic properties of interrogated media. The inverse problem was formulated and solved in the frequency domain and its robustness to noise and geometric constraints was evaluated. The proposed model was validated in vitro with two independent rheology methods on several homogeneous and heterogeneous breast tissue-mimicking phantoms over a broad range of frequencies (up to 400 Hz). Results: Viscoelastic properties matched benchmark rheology methods with discrepancies of 8%–38% for the shear modulus G′ and 9%–67% for the loss modulus G″. The robustness study indicated good estimations of storage and loss moduli (maximum mean errors of 19% on G′ and 32% on G″) for signal-to-noise ratios between 19.5 and 8.5 dB. Larger errors were noticed in the case of biases in lesion dimension and position. Conclusions: The ATSW method revealed that it is possible to estimate the viscoelasticity of biological tissues with torsional shear waves when small biases in lesion geometry exist.

60 APPLIED LIFE SCIENCES↗

DPC Disposal Thermal Scoping Analysis

This is a progress report on thermal modeling for dual-purpose canister (DPCs) direct disposal that covers several available calculation methods and addresses creep and temperature-dependent properties in a salt repository. Three modeling approaches are demonstrated: A semi-analytical calculation method that uses linear solutions with superposition and imaging, to represent a central waste package in a larger array; A finite difference model of coupled thermal creep, implemented in FLAC2D; and An integrated finite difference thermal-hydrologic modeling approach for repositories in different generic host media, implemented in PFLOTRAN. These approaches are at different levels of maturity, and future work is expected to add refinements and establish the best applications for each.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Simulation of the temperature distribution in the selective beam melting process for polymer material

In the present contribution the temperature distribution in the selective beam melting process for polymer materials is simulated to better understand the influence of process parameters on the properties of the produced part. The basis for the developed simulation tool is the nonlinear heat equation including temperature dependent functions for the heat capacity and the heat conduction which were obtained by experimental measurements. The effect of latent heat occurring in the process is also taken into account. The heat equation is discretized in time and space where a Runge-Kutta method of Radau IIA type is used for time integration. An adaptive finite element method is applied for the discretization in space and the model is implemented into the finite element library deal.II. The heat and cooling rate as important process parameters are simulated for different beam velocities. The ability for computing these process parameters makes the simulation tool suited for optimizing the process management of selective beam melting plants.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Incremental ECAP of Tubular Components--FE Simulation

Severe plastic deformation is a method of refining grain structure and improving properties of metallic materials. It is usually carried out on simple billets such as bars, plates and sheets, which are then converted into more complex shapes. One of those shapes, finding applications in many products, is tube. This paper presents an alternative approach to manufacturing tubes with ultrafine grained structure, which is based on direct severe plastic deformation of standard, coarse grained tubes. A new process of incremental equal channel angular pressing is proposed to be used for this purpose. Finite element simulation is used to analyse different process configurations and tool geometry, giving a valuable insight into feasibility of this new approach.

42 ENGINEERING↗

A partially reflecting random walk on spheres algorithm for electrical impedance tomography

In this work, we develop a probabilistic estimator for the voltage-to-current map arising in electrical impedance tomography. This novel so-called partially reflecting random walk on spheres estimator enables Monte Carlo methods to compute the voltage-to-current map in an embarrassingly parallel manner, which is an important issue with regard to the corresponding inverse problem. Our method uses the well-known random walk on spheres algorithm inside subdomains where the diffusion coefficient is constant and employs replacement techniques motivated by finite difference discretization to deal with both mixed boundary conditions and interface transmission conditions. We analyze the global bias and the variance of the new estimator both theoretically and experimentally. Subsequently, the variance of the new estimator is considerably reduced via a novel control variate conditional sampling technique which yields a highly efficient hybrid forward solver coupling probabilistic and deterministic algorithms.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Matrix Methods for Solving Hartree-Fock Equations in Atomic Structure Calculations and Line Broadening

Atomic structure of N-electron atoms is often determined by solving the Hartree-Fock equations, which are a set of integro-differential equations. The integral part of the Hartree-Fock equations treats electron exchange, but the Hartree-Fock equations are not often treated as an integro-differential equation. The exchange term is often approximated as an inhomogeneous or an effective potential so that the Hartree-Fock equations become a set of ordinary differential equations (which can be solved using the usual shooting methods). Because the Hartree-Fock equations are an iterative-refinement method, the inhomogeneous term relies on the previous guess of the wavefunction. In addition, there are numerical complications associated with solving inhomogeneous differential equations. This work uses matrix methods to solve the Hartree-Fock equations as an integro-differential equation. It is well known that a derivative operator can be expressed as a matrix made of finite-difference coefficients; energy eigenvalues and eigenvectors can be obtained by using linear-algebra packages. The integral (exchange) part of the Hartree-Fock equation can be approximated as a sum and written as a matrix. The Hartree-Fock equations can be solved as a matrix that is the sum of the differential and integral matrices. We compare calculations using this method against experiment and standard atomic structure calculations. This matrix method can also be used to solve for free-electron wavefunctions, thus improving how the atoms and free electrons interact. This technique is important for spectral line broadening in two ways: it improves the atomic structure calculations, and it improves the motion of the plasma electrons that collide with the atom.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Highly Accurate Frequency Calculations of Crab Cavities Using the VORPAL Computational Framework

We have applied the Werner-Cary method [J. Comp. Phys. 227, 5200-5214 (2008)] for extracting modes and mode frequencies from time-domain simulations of crab cavities, as are needed for the ILC and the beam delivery system of the LHC. This method for frequency extraction relies on a small number of simulations, and post-processing using the SVD algorithm with Tikhonov regularization. The time-domain simulations were carried out using the VORPAL computational framework, which is based on the eminently scalable finite-difference time-domain algorithm. A validation study was performed on an aluminum model of the 3.9 GHz RF separators built originally at Fermi National Accelerator Laboratory in the US. Comparisons with measurements of the A15 cavity show that this method can provide accuracy to within 0.01% of experimental results after accounting for manufacturing imperfections. To capture the near degeneracies two simulations, requiring in total a few hours on 600 processors were employed. This method has applications across many areas including obtaining MHD spectra from time-domain simulations.

43 PARTICLE ACCELERATORS↗

Mach cones in the quark-gluon plasma: Viscosity, speed of sound, and effects of finite source structure

I use the space-time distribution of energy and momentum deposited by a fast parton traversing a perturbative quark-gluon plasma as a source term for the linearized hydrodynamical equations of the medium. A method of solution for the medium response is presented in detail. Numerical results are given for different values of the shear viscosity to entropy density ratio, {eta}/s, and speed of sound, c{sub s}. Furthermore, I investigate the relevance of finite source structure by expanding the source term up to first order in gradients of a {delta} function centered at the fast parton and comparing the resulting dynamics to that obtained with the full source. It is found that, for the source term used here, the medium response is sensitive to the finite source structure up to distances of several femtometers from the source parton.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗