20&21 EigenVal&EigenVectors

Download as pdf or txt
Download as pdf or txt
You are on page 1of 22

4 Eigenvalue and eigenvector

method3
Consider the system

.. (4.1)

or equivalently
(4.2)
where the matrix

Lecture 20, 08
Lecture 20, 21May
September
2012 2009
44
has constant entries. Eq. (4.1) or (4.2) is called “homoge-
neous first-order linear systems of differential equations with
constant coefficients”. Our aim is to find linearly indepen-
dent solutions on of the system.
By analogy with the scalar case, we look for a solution of the
form
(4.3)
for appropriate vector and scalar

45
Theorem 4.1
Eq. (4.3) provides a nontrivial solution of (4.2) if and
only if and are such that
i.e. ( (4.4)
Proof

Since the above implies that

Definition 4.2
A scalar is called an eigenvalue of the matrix if there
exists a nonzero vector such Eq. (4.4) holds such a vector

46
is called an eigenvector associated with the eigenvalue

47
How to find eigenvalues?
Given , the system (4.4) has a nontrivial solution iff

.. (4.5)

Eq. (4.5) in the unknown is called the characteristic


equation of the matrix Its roots are the eigenvalues of the
matrix As a polynomial of degree , Eq. (45) has roots
(real or complex) counted with their multiplicities.

48
How to solve the system (4.1)?
1. Solve the characteristic equation (4.5) to find the
eigenvalues of
2. “Try” to find linearly independent associated
eigenvectors
3. Step (2) is not always possible but should it be,
we obtain linearly independent solutions

If step (2) does not work, we must find some other


ways.
4. The general solution is then the -parameter family
49
of vector functions:

50
Connection with linear algebra
Assume that the matrix is diagonalisable i.e
diag( (4.6)
where

.. (4.7)

is the transition matrix. Make the change of dependent


variable:
(4.8)
Then

51
The latter is an uncoupled system with general solution

Thus, from (4.7) and (4.8),

52
Theorem 4.3
If are real and distinct eigenvalues of
and are associated eigenvectors that are linearly
independent, then

are linearly independent solutions of (4.2) and its general


solution is

Proof.

53
by linear independence of
Example 4.4
Find the general solution of the system

Details in class

54
Theorem 4.54
If

is a complex eigenvalue of with an associated eigenvector

then :
The vector function

is a solution of the system (4.2) with complex-


Lecture21,
Lecture 21,10
21May
Septemer
2012 2009
55
valued components
The vector functions

and

are linearly independent solutions of (4.2) that


have real-valued components.

56
Proof.
The first bullet follows from Theorem 4.1, which also
implies that the real and imaginary parts of are solutions
with real-valued components. Let us show that these two
solutions are linearly independent.

as there are more equations than the unknowns.

57
Remark 4.6
Do not memorize the formulae in the second bullet of
Theorem 4.5. Rather memorize the formula in the first bullet
and take its real and imaginary parts.
Homework 4.7
Write down and prove Theorem 4.5 by using the fact that
the conjugate number is also an eigenvalue of

58
Example 4.8
Find the solution of the IVP

Details in class

59
Application to mixture problems
Consider Example 2 (page 371) in textbook that leads to
the system

with initial conditions

Given the triangular structure of the matrix we


get the characteristic equation

So the eigenvalues of are:

60
Eigenvectors associated with

)
i.e.

or

61
which implies that

The space of eigenvectors associated with


is

Taking yields the eigenvector

Homework. In a similar manner, find

62
as eigenvectors associated with and
respectively.
By Theorem 4.3

are linearly independent solutions of the system.

63
The general solution is

Find such that

i.e.

64
We find

Thus

65

You might also like