Spectral Methods in the Presence of Discontinuities

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 L 2 -norm. However, for non-smooth problems, convergence is significantly worse—the L 2 -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. This is an important first step towards using these techniques for simulations of realistic systems.

[1]  Alex Solomonoff,et al.  On the Gibbs phenomenon I: recovering exponential accuracy from the Fourier partial sum of a nonperiodic analytic function , 1992 .

[2]  Anne Gelb,et al.  Detection of Edges in Spectral Data III—Refinement of the Concentration Method , 2008, J. Sci. Comput..

[3]  Hervé Vandeven,et al.  Family of spectral filters for discontinuous problems , 1991 .

[4]  Anne Gelb,et al.  Spectral Viscosity for Shallow Water Equations in Spherical Geometry , 2001 .

[5]  Andreas Meister,et al.  Application of spectral filtering to discontinuous Galerkin methods on triangulations , 2012 .

[6]  Anne Gelb,et al.  Enhanced spectral viscosity approximations for conservation laws , 2000 .

[7]  Eric Jones,et al.  SciPy: Open Source Scientific Tools for Python , 2001 .

[8]  John D. Hunter,et al.  Matplotlib: A 2D Graphics Environment , 2007, Computing in Science & Engineering.

[9]  Eitan Tadmor,et al.  Adaptive Mollifiers for High Resolution Recovery of Piecewise Smooth Data from its Spectral Information , 2001, Found. Comput. Math..

[10]  J. Preston.  Ξ-filters , 1983 .

[11]  John P. Boyd,et al.  Trouble with Gegenbauer reconstruction for defeating Gibbs' phenomenon: Runge phenomenon in the diagonal limit of Gegenbauer polynomial approximations , 2005 .

[12]  A. Michelson,et al.  Fourier's Series , 1898, Nature.

[13]  Jan S. Hesthaven,et al.  Discontinuous Galerkin method for computing gravitational waveforms from extreme mass ratio binaries , 2009, 0902.1287.

[14]  S. Nissanke,et al.  Binary-black-hole merger: symmetry and the spin expansion. , 2007, Physical review letters.

[15]  J. Hesthaven,et al.  Nodal Discontinuous Galerkin Methods: Algorithms, Analysis, and Applications , 2007 .

[16]  Jared Tanner,et al.  Optimal filter and mollifier for piecewise smooth spectral data , 2006, Math. Comput..

[17]  Anne Gelb,et al.  Determining Analyticity for Parameter Optimization of the Gegenbauer Reconstruction Method , 2005, SIAM J. Sci. Comput..

[18]  Knut S. Eckhoff.  On discontinuous solutions of hyperbolic equations , 1994 .

[19]  Lawrence E. Kidder,et al.  High-accuracy waveforms for binary black hole inspiral, merger, and ringdown , 2008, 0810.1767.

[20]  J. Gibbs.  Fourier's Series , 1898, Nature.

[21]  Eitan Tadmor,et al.  Filters, mollifiers and the computation of the Gibbs phenomenon , 2007, Acta Numerica.

[22]  Anne Gelb,et al.  Detection of Edges in Spectral Data , 1999 .

[23]  G. vanRossum.  Python reference manual , 1995 .

[24]  A. Quarteroni,et al.  Approximation results for orthogonal polynomials in Sobolev spaces , 1982 .

[25]  P. Cochat,et al.  Et al , 2008, Archives de pediatrie : organe officiel de la Societe francaise de pediatrie.

[26]  Dick Wick Hall,et al.  Elementary Real Analysis , 1971 .

[27]  D. Gottlieb,et al.  A new method of imposing boundary conditions in pseudospectral approximations of hyperbolic equations , 1988 .

[28]  D. Gottlieb,et al.  Spectral methods for hyperbolic problems , 2001 .

[29]  Nathaniel R. Morgan,et al.  A Lagrangian discontinuous Galerkin hydrodynamic method , 2018 .

[30]  Cornelius Lanczos,et al.  Discourse on Fourier series , 1966 .

[31]  E. Hewitt,et al.  The Gibbs-Wilbraham phenomenon: An episode in fourier analysis , 1979 .

[32]  David Levin,et al.  High order approximation to non-smooth multivariate functions , 2018, Comput. Aided Geom. Des..

[33]  V.-T. Nguyen,et al.  A discontinuous Galerkin front tracking method for two-phase flows with surface tension , 2008 .

[34]  Anne Gelb,et al.  Robust reprojection methods for the resolution of the Gibbs phenomenon , 2006 .

[35]  A. Love,et al.  Fourier's Series , 1898, Nature.

[36]  A. Stroud,et al.  Approximate Calculation of Integrals , 1962 .

[37]  David Levin,et al.  Approximating piecewise-smooth functions , 2010 .

[38]  Anne Gelb,et al.  Detection of Edges in Spectral Data II. Nonlinear Enhancement , 2000, SIAM J. Numer. Anal..

[39]  Eitan Tadmor,et al.  Shock capturing by the spectral viscosity method , 1990 .

[40]  Gaël Varoquaux,et al.  The NumPy Array: A Structure for Efficient Numerical Computation , 2011, Computing in Science & Engineering.

[41]  Jae-Hun Jung,et al.  A Review of David Gottlieb's Work on the Resolution of the Gibbs Phenomenon , 2011 .

[42]  David Gottlieb,et al.  CONVERGENCE RESULTS FOR PSEUDOSPECTRAL APPROXIMATIONS OF HYPERBOLIC SYSTEMS BY A PENALTY-TYPE BOUNDARY TREATMENT , 1991 .

[43]  J. Novak,et al.  Spectral Methods for Numerical Relativity , 2007, Living reviews in relativity.

[44]  Eitan Tadmor,et al.  Recovering Pointwise Values of Discontinuous Data within Spectral Accuracy , 1985 .