Solving Polynomial Eigenvalue Problem

Click For Summary
SUMMARY

The discussion focuses on solving the polynomial eigenvalue problem defined by the equation M λ² + Φ λ + K, where K is specified as [[5.92, -0.9837], [-0.3381, 109.94]] and M as [[1, 0], [0, 1]]. The user employs the MATLAB function polyeig to compute eigenvalues and eigenvectors, but encounters unexpected results with the first eigenvalue outputting a non-zero imaginary component. The issue is identified as a potential misordering of arguments in the polyeig function, which is critical for accurate computations.

PREREQUISITES
  • Understanding of polynomial eigenvalue problems
  • Familiarity with MATLAB programming
  • Knowledge of the polyeig function in MATLAB
  • Basic linear algebra concepts, including eigenvalues and eigenvectors
NEXT STEPS
  • Review the MATLAB documentation for polyeig to ensure proper argument ordering
  • Explore examples of polynomial eigenvalue problems in MATLAB
  • Learn about the implications of complex eigenvalues in physical systems
  • Investigate condition numbers and their impact on numerical stability in eigenvalue computations
USEFUL FOR

Mathematicians, engineers, and researchers working on eigenvalue problems, particularly those utilizing MATLAB for numerical analysis and simulations.

member 428835
Hi PF!

I'm trying to solve the polynomial eigenvalue problem ##M \lambda^2 + \Phi \lambda + K## such that
Code:
K = [5.92 -.9837;-0.3381 109.94];
I*[14.3 24.04;24.04 40.4];
M = [1 0;0 1];
[f lambda cond] = polyeig(M,Phi,K)
I verify the output of the first eigenvalue via
Code:
(M*lambda(1)^2 + Phi*lambda(1) + K)*f(:,1)
and the output has zero real component but non-zero imaginary: ##[44i,88i]##. Any ideas what's happening? Condition number is 5, which is evidently sufficiently low to output accurate answers?
 
Physics news on Phys.org
I think you are giving polyeig the arguments in the wrong order.

Jason
 
  • Like
Likes   Reactions: member 428835
jasonRF said:
I think you are giving polyeig the arguments in the wrong order.

Jason
Thanks!
:doh:
 

Similar threads

  • · Replies 1 ·
Replies
1
Views
3K
  • · Replies 13 ·
Replies
13
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 4 ·
Replies
4
Views
3K
Replies
11
Views
6K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 8 ·
Replies
8
Views
2K
Replies
13
Views
2K
  • · Replies 8 ·
Replies
8
Views
2K