Scribd
Upload
429 views3 pages

Cayley Hamilton Examples

The document provides three examples of using the Cayley-Hamilton theorem to evaluate functions of matrices. In the first example, it shows how to use the theorem to evaluate the polynomial function f(A) = A^5 - I for a 3x3 matrix A. In the second example, it evaluates the exponential function f(A) = e^At. The third example calculates the inverse of the matrix sI - A.

Uploaded by

hardcore85
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
429 views3 pages

Cayley Hamilton Examples

The document provides three examples of using the Cayley-Hamilton theorem to evaluate functions of matrices. In the first example, it shows how to use the theorem to evaluate the polynomial function f(A) = A^5 - I for a 3x3 matrix A. In the second example, it evaluates the exponential function f(A) = e^At. The third example calculates the inverse of the matrix sI - A.

Uploaded by

hardcore85
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

ECE 602 Lumped Systems Theory September 12, 2013 1

ECE 602 Lecture Notes:


Cayley-Hamilton Examples
The Cayley Hamilton Theorem states that a square n n matrix A satises its own
characteristic equation. Thus, we can express A
n
in terms of a nite set of lower powers
of A. This fact leads to a simple way of calculating the value of a function evaluated at
the matrix. This method is given in Theorem 3.5 of the textbook
1
. Here we give a couple
of examples.
We will consider polynomial functions f(A) and the exponential function e
At
. Well work
with the matrix
A =

1 1 0
0 1 0
0 0 1

, (1)
which has the convenient property that
A
k
=

1 k 0
0 1 0
0 0 1

. (2)
The characteristic equation of A can be obtained easily by noting the second and third
rows each have only a single nonzero entry. Using the second we have
() = (s 1)
3
, (3)
so the matrix has a single eigenvalue 1 with multiplicity three.
Because the matrix is 3 3 we will use the polynomial
h() :=
0
+
1
+
2

2
(4)
in our calculations. We select the notation f
(k)
() := d
k
f/d
k
.
Example 1 Consider the polynomial f() =
5
1.
We obtain three linear equations in three unknowns as follows. Because the eigen-
value of A has multiplicity three we must use two derivatives of f and h. Our
equations are
f() =
5
1 h() =
0
+
1
+
2

2
(5)
f
(1)
() = 5
4
h
(1)
() =
1
+ 2
2
(6)
f
(2)
() = 20
3
h
(2)
() = 2
2
. (7)
1
Chen, C.-T. Linear System Theory and Design, New York, Oxford University Press, 1999.
ECE 602 Lumped Systems Theory September 12, 2013 2
We obtain
f(1) = 0 =
0
+
1
+
2
= h(1) (8)
f
(1)
(1) = 5 =
1
+ 2
2
= h
(1)
(1) (9)
f
(2)
(1) = 20 = 2
2
= h
(2)
(1). (10)
Solving for the s we nd

2
= 10,
1
= 15,
0
= 5. (11)
Thus
f(A) =
0
A
0
+
1
A
1
+
2
A
2
(12)
= 5I 15A+ 10A
2
(13)
=

0 5 0
0 0 0
0 0 0

(14)
= A
5
I. (15)
Example 2 Consider the function f() = e
t
.
We obtain three linear equations in three unknowns as follows. Because the eigen-
value of A has multiplicity three we must use two derivatives of f and h. Remem-
bering that we are taking derivatives with respect to , our equations are
f() = e
t
h() =
0
+
1
+
2

2
(16)
f
(1)
() = te
t
h
(1)
() =
1
+ 2
2
(17)
f
(2)
() = t
2
e
t
h
(2)
() = 2
2
. (18)
We obtain
f(1) = e
t
=
0
+
1
+
2
= h(1) (19)
f
(1)
(1) = te
t
=
1
+ 2
2
= h
(1)
(1) (20)
f
(2)
(1) = t
2
e
t
/2 = 2
2
= h
(2)
(1). (21)
Solving for the s we nd

2
= t
2
e
t
/2,
1
= te
t
t
2
e
t
,
0
= e
t
te
t
+ t
2
e
t
/2. (22)
Thus
f(A) =
0
A
0
+
1
A
1
+
2
A
2
(23)
= e
t

1 t + t
2
/2

I + e
t

t t
2

A+ e
t

t
2
/2

A
2
(24)
=

e
t
te
t
0
0 e
t
0
0 0 e
t

. (25)
ECE 602 Lumped Systems Theory September 12, 2013 3
Example 3 Computation of (sI A)
1
for
A =

0 1
1 2

using equivalence on the spectrum of A.


First lets nd the spectrum of A.
|sI A| =

s 1
1 s + 2

= s(s + 2) (1) = s
2
+ 2s + 1 = (s + 1)
2
so A has eigenvalue = 1 with multiplicity 2.
Next, let f() = (s )
1
and, as always, h() =
0
+
1
.
Then f
(1)
() = (1)(s )
2
and h
(1)
() =
1
.
Substituting the eigenvalue for yields

1
= (s + 1)
2
(26)

0
= (s + 1)
1
(1)(s + 1)
2
= (s + 2)(s + 1)
2
(27)
Thus h() = (s + 2)(s + 1)
2
+ (s + 1)
2
and substituting A for yields
f(A) = (sI A)
1
(28)
= (s + 2)(s + 1)
2
I + (s + 1)
2
A (29)
=

s+2
(s+1)
2
1
(s+1)
2
1
(s+1)
2
(s+22)
(s+1)
2

. (30)

You might also like