UW Applied Mathematics

Computing Spectra of Linear Operators Using Finite Differences

J. Nathan Kutz
Department of Applied Mathematics University of Washington Seattle, WA 98195-2420 Email: (kutz@amath.washington.edu)
Stability Workshop, Seattle, September 6-8, 2006

Introduction: Spectral Stability UW Applied Mathematics

Consider the nonlinear PDE evolution

Equilibrium solution

Linear stability

perturbation

B. Deconinck and J. N. Kutz, J. Comp. Phys. (2006)

Associated Eigenvalue Problem UW Applied Mathematics

Separation of variables

Eigenvalue problem

Linear stability

Re( ) 0 Re( ) > 0

spectrally stable spectrally unstable

Finite Differences and Taylor Series UW Applied Mathematics

Taylor expand

Add

approximation

error

slope formula with error

Higher-Order Accuracy UW Applied Mathematics

Taylor expand again

8 x (first) and subtract

approximation error

slope formula with improved error

Finite Difference Tables UW Applied Mathematics

neighboring points determine accuracy

Forward and Backward Differences UW Applied Mathematics

Required for incorporating boundary conditions

asymmetric neighboring points

Numerical Round-Off UW Applied Mathematics

Consider the error in approximating the first derivative

The error includes round-off and truncation

round-off truncation

Assume round-off

and minimum at

round-off error dominates below

Boundary Conditions: Pinned UW Applied Mathematics

pinned boundaries tri-diagonal matrix structure

Boundary Conditions: Periodic UW Applied Mathematics

periodic boundaries tri-diagonal matrix structure with corners

Boundary Conditions: No Flux UW Applied Mathematics

no flux condition

no longer symmetry matrix

General Boundaries UW Applied Mathematics

General (Sturm-Liouville) boundary conditions

Difficult to incorporate into matrix structure shooting methods relaxation methods

no longer symmetry matrix

Algorithm UW Applied Mathematics

choose domain length and discretization size construct linear operator implement boundary conditions use eigenvalue/eigenvector solver: O(N3) (or shooting/relaxation methods) construct eigenfunctions

Easily extends to vectors and higher dimensions

Example: Mathieu Equation UW Applied Mathematics

Classic Example

Operator

is self-adjoint (real spectrum)

compute with matlab, maple, mathematica, or homemade code

Spectrum for Mathieu Equation UW Applied Mathematics

a
a is eigenvalue

Computing the Ground State UW Applied Mathematics

Convergence study and CPU time (q=2)

0.43 sec 0.01 sec

1 hour 0.77 sec

2.5 min

beyond Matlab7s ability

What about band-gap structure increase domain length Floquet theory

Calculating the Bands: Domain Length UW Applied Mathematics

Increase the domain length (q=2)

doubling gives 8x computational increase

traditional way: very costly for recovering bands

Calculating the Bands: Floquet Theory UW Applied Mathematics

Make use of Floquet (Bloch) theory
larger period solutions

with Floquet (characteristic) exponents keep fixed domain discretize solve D O(N3) equations

Implementing Floquet Theory UW Applied Mathematics

Floquet theory modifies matrix corners

with Floquet slices

Floquet Theory vs. Domain Length UW Applied Mathematics

Compare methods for computing band (q=2)

band density

9 min 8 sec

1 min

10 min

beyond Matlab7s ability

Use Floquet Theory!

Example: Periodic NLS UW Applied Mathematics

Consider the system

with

and Jacobi sine function

Spectrum of Periodic NLS UW Applied Mathematics

k=0

k=0.7

k=0.9

k=1

Importance of Floquet Slicing UW Applied Mathematics

0

=0

Accuracy and Convergence UW Applied Mathematics

Example: 2D Mathieu Equation UW Applied Mathematics

Consider

with

Operation Count: O((N2)3)=O(N6)

Laplacian in 2D UW Applied Mathematics

Consider

Discretize:

Let

must stack 2D data: periodic boundaries add structure

Laplacian in 2D UW Applied Mathematics

nx=ny=4

Laplacian in 2D UW Applied Mathematics

nx=ny=4

Matlab easily builds 2D Laplacian

Band Gap Structure UW Applied Mathematics

A=-0.3

A=0

A=1

First three band-gap structures

Quasi-momentum representation

2D Dominant Eigenfunctions UW Applied Mathematics

First three eigenfunctions for x = y=0

2D Eigenfunctions UW Applied Mathematics

First three eigenfunctions for x = y=1/4

Summary and Conclusions UW Applied Mathematics

simple, simple, simple boundary conditions at edge of matrices eigenvalue solvers make use of sparse structure Floquet theory for resolution of bands costly/impractical for 2D-3D problems

B. Deconinck and J. N. Kutz, J. Comp. Phys. (2006)