Applied Numerical Methods

with MATLAB®
for Engineers and Scientists
4th Edition
Steven C. Chapra

PowerPoints organized by Dr. Michael R. Gustafson II, Duke University and

Prof. Steve Chapra, Tufts University

©McGraw-Hill Education. All rights reserved. Authorized only for instructor use in the classroom. No reproduction or further distribution permitted without the prior written consent of McGraw-Hill Education.
 Chapter 13

Transformations and
Eigenvalues

©McGraw-Hill Education. All rights reserved. Authorized only for instructor use in the classroom. No reproduction or further distribution permitted without the prior written consent of McGraw-Hill Education.
• Dynamics of Three Story Building

©McGraw-Hill Education.
• Principle Modes of Vibration, 2

©McGraw-Hill Education.
Some Applications of Eigenvalue Problems
Find the principal directions

 y1   5 3   x1 

n
x12  x 22 1, y   

io
  

ct
pr
 y 2   3 5  x 2 

ire
in
ci

ld
pa

pa
 1

ld

ci
y  A x  x  1 8, x1  

ire

in
pr
ct
 1

io
n
 1
2 2, x 2  
  1

Eigenvector ~ direction
 Approaches
 Find the characteristic polynomial
– Leverrier’s Method
 Find the largest or smallest eigenvalue
– Direct Power Method
– Inverse Power Method
 Find all the eigenvalues
– Jacobi’s Method
– Householder’s Method
– QR Method
– Danislevsky’s Method

6
• The Power Method
Iterative method to compute the largest eigenvalue and its associated
eigenvector.

Simple Algorithm:

1. i=1
2. Start with x1=ones(n,1)
3. i=i+1
4. Find v=A . xi
5. Find MAX (v)
6. Xi=v/MAX(v)
7. Repeat from step 3 until convergence

©McGraw-Hill Education.
• The Power Method
Iterative method to compute the largest eigenvalue and its associated
eigenvector.

Simple Algorithm:

function [eval, evect] = powereig(A,es,maxit)

n=length(A);
evect=ones(n,1);eval=1;iter=0;ea=100; %initialize
while(1)
evalold=eval; %save old eigenvalue value
evect=A*evect; %determine eigenvector as [A]*{x)
eval=max(abs(evect)); %determine new eigenvalue
evect=evect./eval; %normalize eigenvector to eigenvalue
iter=iter+1;
if eval~=0, ea = abs((eval-evalold)/eval)*100; end
if ea<=es | iter >= maxit,break,end
end
©McGraw-Hill Education.
• Example: The Power Method, 1
First iteration:

Second iteration:

©McGraw-Hill Education.
• Example: The Power Method, 2
Third iteration:

Fourth iteration:

©McGraw-Hill Education.
• Example: The Power Method, 3
Fifth iteration:

The process can be continued to determine the largest

eigenvalue (= 66.4) with the associated eigenvector
[-0.7 1 -0.7]

Note that the smallest eigenvalue and its

associated eigenvector can be determined by
applying the power method to the inverse of A
©McGraw-Hill Education.
• Determining Eigenvalues &
Eigenvectors with MATLAB
>> A = [10 -5;-5 10]

A =
10 -5
-5 10

>> p = poly(A)

>> Ev=roots(p)

©McGraw-Hill Education.
• Determining Eigenvalues &
Eigenvectors with MATLAB
>> A = [10 -5;-5 10]

A =
10 -5
-5 10

>> [v,lambda] = eig(A)

v =
-0.7071 -0.7071
-0.7071 0.7071
lambda =
5 0
0 15
©McGraw-Hill Education.
• Determining Eigenvalues &
Eigenvectors with MATLAB

A =
5/4 0
0 3/4

-Use eig command on A to find

eigenvalues and eigenvectors

-verify by running eigshow.m for A

©McGraw-Hill Education.
 DIAGONALIZATION OF A MATRIX

Theorem: If an n × n matrix A has a basis of eigenvectors, then

D = X−1AX

is diagonal, with the eigenvalues of A as the entries on the main diagonal. Here
X is the matrix with these eigenvectors as column vectors.

Also:

Dm = X−1AmX (m = 2, 3, … ).

• 1
©McGraw-Hill Education. 5
 EXAMPLE

Diagonalize:
7.3 0.2 −3.7

[
𝐀 = −11.5
17.7
1.0
1.8
5.5
− 9.3 ]
Solution:

− 0.7 0.2 0.3 −3 −4 0 3 0 0

𝐃= 𝐗
−1

[
𝐀𝐗 = − 1.3
0.8
− 0.2
0.2
0.7
− 0.2 ][ 9
−3
4
− 12 ][
0 = 0
0 0
−4
0
0
0 ]
• 1
©McGraw-Hill Education. 6