Spectral methods in the presence of discontinuities

doi:10.1016/j.jcp.2019.03.048

Title: Spectral methods in the presence of discontinuities

Full Record
Other Related Research

Abstract

Spectral methods provide an elegant and efficient way of numerically solving differential equations of all kinds. For smooth problems, truncation error for spectral methods vanishes exponentially in the infinity norm and L₂-norm. However, for non-smooth problems, convergence is significantly worse—the L₂-norm of the error for a discontinuous problem will converge at a sub-linear rate and the infinity norm will not converge at all. We explore and improve upon a post-processing technique—optimally convergent mollifiers—to recover exponential convergence from a poorly-converging spectral reconstruction of non-smooth data. Furthermore this is an important first step towards using these techniques for simulations of realistic systems.

Authors:

^[1];

^[2];

^[3]

Univ. of Cambridge, Cambridge (United Kingdom); Perimeter Inst. for Theoretical Physics, Waterloo, ON (Canada)
Los Alamos National Lab. (LANL), Los Alamos, NM (United States)
Perimeter Inst. for Theoretical Physics, Waterloo, ON (Canada); Univ. of Waterloo, Waterloo, ON (Canada); Louisiana State Univ., Baton Rouge, LA (United States)

Publication Date:: 2019-04-11

Research Org.:: Los Alamos National Lab. (LANL), Los Alamos, NM (United States)

Sponsoring Org.:: USDOE Office of Science (SC)

OSTI Identifier:: 1508545

Alternate Identifier(s):: OSTI ID: 1547523

Report Number(s):: LA-UR-17-31492
Journal ID: ISSN 0021-9991

Grant/Contract Number:: 89233218CNA000001; 20170508DR; SC0018297

Resource Type:: Accepted Manuscript

Journal Name:: Journal of Computational Physics

Additional Journal Information:: Journal Volume: 390; Journal ID: ISSN 0021-9991

Publisher:: Elsevier

Country of Publication:: United States

Language:: English

Subject:: 71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSICS; Spectral method; Numerical analysis; Gibbs' phenomenon; Shock handling; Discontinuous functions; Partial differential equations

Citation Formats


                    Piotrowska, Joanna, Miller, Jonah Maxwell, and Schnetter, Erik. Spectral methods in the presence of discontinuities.  United States: N. p., 2019. 
        Web.  doi:10.1016/j.jcp.2019.03.048.

Copy to clipboard


                    Piotrowska, Joanna, Miller, Jonah Maxwell, & Schnetter, Erik. Spectral methods in the presence of discontinuities.  United States.  doi:10.1016/j.jcp.2019.03.048.

Copy to clipboard


                    Piotrowska, Joanna, Miller, Jonah Maxwell, and Schnetter, Erik. Thu .  
        "Spectral methods in the presence of discontinuities".  United States.  doi:10.1016/j.jcp.2019.03.048.  https://www.osti.gov/servlets/purl/1508545.

Copy to clipboard


                    
@article{osti_1508545,

  title        = {Spectral methods in the presence of discontinuities},

  author       = {Piotrowska, Joanna and Miller, Jonah Maxwell and Schnetter, Erik},

  abstractNote = {Spectral methods provide an elegant and efficient way of numerically solving differential equations of all kinds. For smooth problems, truncation error for spectral methods vanishes exponentially in the infinity norm and L2-norm. However, for non-smooth problems, convergence is significantly worse—the L2-norm of the error for a discontinuous problem will converge at a sub-linear rate and the infinity norm will not converge at all. We explore and improve upon a post-processing technique—optimally convergent mollifiers—to recover exponential convergence from a poorly-converging spectral reconstruction of non-smooth data. Furthermore this is an important first step towards using these techniques for simulations of realistic systems.},

  doi          = {10.1016/j.jcp.2019.03.048},

  journal      = {Journal of Computational Physics},

  number       = ,

  volume       = 390,

  place        = {United States},

  year         = {2019},

  month        = {4}

}

Copy to clipboard

Journal Article:

Free Publicly Available Full Text

Accepted Manuscript (Publisher)

Accepted Manuscript (DOE)

Publisher's Version of Record

DOI: 10.1016/j.jcp.2019.03.048

Other availability

Search WorldCat to find libraries that may hold this journal

Save / Share:

Export Metadata

Save to My Library

Similar Records in DOE PAGES and OSTI.GOV collections:

Error Analysis of Variations on Larsen's Benchmark Problem

Conference Azmy, YY

Error norms for three variants of Larsen's benchmark problem are evaluated using three numerical methods for solving the discrete ordinates approximation of the neutron transport equation in multidimensional Cartesian geometry. The three variants of Larsen's test problem are concerned with the incoming flux boundary conditions: unit incoming flux on the left and bottom edges (Larsen's configuration); unit, incoming flux only on the left edge; unit incoming flux only on the bottom edge. The three methods considered are the Diamond Difference (DD) method, and the constant-approximation versions of the Arbitrarily High Order Transport method of the Nodal type (AHOT-N), and ofmore »« less
Full Text Available
Response matrix properties and convergence implications for an interface-current nodal formulation

Journal Article Yang, W S - Nuclear Science and Engineering

An analytic study was performed of the properties and the associated convergence implications of the response matrix equations derived via the widely used nodal expansion method. By using the DIF3D nodal formulation in hexagonal-z geometry as a concrete example, an analytic expression for the response matrix is first derived by using the hexagonal prism symmetry transformations. The spectral radius of the local response matrix is shown to be always <1. The l{sub 2}-norm of the response matrix is shown to be <1 in two-dimensional problems but not always <1 in three-dimensional problems. The elements of the response matrix are shownmore »« less
Accuracy and convergence properties of multiple-scattering theory in three dimensions

Journal Article Butler, W H ; Zhang, X - Physical Review, B: Condensed Matter; (United States)

We present a numerical and analytical study of the accuracy and convergence properties of multiple-scattering theory (MST) in three dimensions. The convergence with respect to the number of angular momentum states, {ital l}{sub max}, of the solutions to the three-dimensional MST equations for two muffin-tin potentials is studied analytically and by means of numerical calculations. The rate of convergence appears to be a universal quantity which depends, in the limit of {ital l}{sub max}{r arrow}{infinity}, only on the separation between the scatterers relative to their radii. No evidence of error in the energy, the wave function, or the derivative ofmore »« less
DOI: 10.1103/PhysRevB.44.969
Transport in the spatially tempered, fractional Fokker-Planck equation

Journal Article Kullberg, A ; Del-Castillo-Negrete, Diego B - Journal of Physics A: Mathematical and Theoretical

A study of truncated Levy flights in super-diffusive transport in the presence of an external potential is presented. The study is based on the spatially tempered, fractional Fokker-Planck (TFFP) equation in which the fractional diffusion operator is replaced by a tempered fractional diffusion (TFD) operator. We focus on harmonic (quadratic) potentials and periodic potentials with broken spatial symmetry. The main objective is to study the dependence of the steady-state probability density function (PDF), and the current (in the case of periodic potentials) on the level of tempering, lambda, and on the order of the fractional derivative in space, alpha. Anmore »« less
DOI: 10.1088/1751-8113/45/25/255101
General criteria for assessing the accuracy of approximate wave functions and their densities

Journal Article Nunez, M A - International Journal of Quantum Chemistry

By means of examples, Loewdin showed that L{sub 2} convergence of approximate wave functions {psi}{sub n} to the exact {psi} using the single limit lim{sub n{r_arrow}{infinity}}{psi}{sub n}, A {psi}{sub n} = ({psi}, A{psi}) is not sufficient to compute accurate expectation values. It is shown that L{sub 2} convergence is indeed a sufficient condition to compute accurate expectation values using iterated limits lim{sub m-x}lim{sub n-{infinity}}({psi}{sub n}A{sub psi}{sub m}) = ({psi}, A{psi}) instead of a single limit. Practical conditions that guarantee the stability of single-limit calculations are given. It is also shown that the L{sub 2} convergence of wave functions implies themore »« less
DOI: 10.1002/qua.560530106

Similar Records