Escolar Documentos
Profissional Documentos
Cultura Documentos
A = [a1 a2 an]
Here only a single subscript is used to denote an element, since the row
subscript is always understood to be equal to 1, that is, a1 = a11, a2 = a12,
and so on.
612
A.1 BASIC DEFINITIONS AND TYPES OF MATRICES 613
a1
a2
A = D T
o
am
a11 0 0
A = C0 a22 0 S
0 0 a33
1 0 0
I = C0 1 0S
0 0 1
3 5 2
A = C5 -1 4S
2 4 8
614 APPENDIX A M AT R I X A L G E B R A FOR S T R U C T U R A L A N A LY S I S
A = c d B = c d
2 6 2 6
4 -3 4 -3
then A = B.
-5
A = c d B = c d
6 7 8
2 -1 1 4
then
-1
A + B = c d A - B = c d
A 1 15 11
3 3 1 -5
A = c d
4 1
k = -6
6 -2
then
-24 -6
kA = c d
-36 12
A = c d B = c d
a11 a12 b11 b12 b13
(A1)
a21 a22 b21 b22 b23
then AB can be determined since A has two columns and B has two rows.
Notice, however, that BA is not possible. Why?
A.2 MATRIX OPERATIONS 615
A B = C
1m * n21n * q2 1m * q2
The elements of matrix C are found using the elements aij of A and bij
of B as follows:
n
cij = a aikbkj (A2)
k=1
2
A = c d
2 4 3
B = C6S
-1 6 1 A
7
Thus
C = c d
49
41
616 APPENDIX A M AT R I X A L G E B R A FOR S T R U C T U R A L A N A LY S I S
5 3
B = c d
2 7
A = C 4 1S
-3 4
-2 8
Here again the product C = AB can be found since A has two columns
and B has two rows. The resulting matrix C will have three rows and two
columns. The elements are obtained as follows:
1 47
A
C = C 5 32 S
-28 18
Notice also that BA does not exist, since written in this manner the
matrices are nonconformable.
The following rules apply to matrix multiplication.
1. In general the product of two matrices is not commutative:
AB Z BA (A3)
A1B + C2 = AB + AC (A4)
Then
1A + B2T = AT + BT (A6)
A = c d B = c d
6 2 4 3
1 -3 2 5
A
Then, by Eq. A8,
3 T
ac dc db = c dc d
6 2 4 4 2 6 1
1 -3 2 5 3 5 2 -3
28 T -2
ac db = c d
28 28
-2 -12 28 -12
-2 -2
c d = c d
28 28
28 -12 28 -12
A21 = c d A22 = c d
a21 a22 a23 a24
a31 a32 a33 a34
618 APPENDIX A M AT R I X A L G E B R A FOR S T R U C T U R A L A N A LY S I S
-7 -4
c d + c d
8 4
1 0
-4 2 -35 - 20
AB = D T = C - 39 -18 S
[12 18] + [56 32] 68 50
A.3 Determinants
In the next section we will discuss how to invert a matrix. Since this
operation requires an evaluation of the determinant of the matrix, we will
now discuss some of the basic properties of determinants.
A determinant is a square array of numbers enclosed within vertical
bars. For example, an nth-order determinant, having n rows and n
columns, is
a11 a12 a1n
a21 a22 a2n 4
A = 4 (A9)
o
an1 an2 ann
A.3 DETERMINANTS 619
C11 = 1-121 + 1 ` ` = ` 22 `
a22 a23 a a23
a32 a33 a32 a33
= 1-121 + 2 ` 21 ` = - ` 21 `
a a23 a a23
C12
a31 a33 a31 a33
= 1-121 + 3 ` 21 ` = ` 21 `
a a22 a a22
C13
a31 a32 a31 a32
A
Laplaces expansion for a determinant of order n, Eq. A9, states that
the numerical value represented by the determinant is equal to the sum
of the products of the elements of any row or column and their respective
cofactors, i.e.,
D = ai1Ci1 + ai2Ci2 + + ainCin 1i = 1, 2, , or n2
or (A11)
D = a1jC1j + a2jC2j + + anjCnj 1j = 1, 2, , or n2
For application, it is seen that due to the cofactors the number D is defined
in terms of n determinants (cofactors) of order n - 1 each. These
determinants can each be reevaluated using the same formula, whereby
one must then evaluate 1n - 12 determinants of order 1n - 22, and so
on. The process of evaluation continues until the remaining determinants
to be evaluated reduce to the second order, whereby the cofactors of the
elements are single elements of D. Consider, for example, the following
second-order determinant
D = ` `
3 5
-1 2
We can evaluate D along the top row of elements, which yields
D = 31 -121 + 1122 + 51-121 + 21- 12 = 11
Or, for example, using the second column of elements, we have
D = 51-121 + 21-12 + 21-122 + 2132 = 11
620 APPENDIX A M AT R I X A L G E B R A FOR S T R U C T U R A L A N A LY S I S
Rather than using Eqs. A11, it is perhaps easier to realize that the
evaluation of a second-order determinant can be performed by
multiplying the elements of the diagonal, from top left down to right, and
subtract from this the product of the elements from top right down to
left, i.e., follow the arrow,
D = ` ` = 3122 - 51-12 = 11
3 5
-1 2
N
Consider next the third-order determinant
1 3 -1
D = 3 4 2 63
-1 0 2
Using Eq. A11, we can evaluate D using the elements either along the
top row or the first column, that is
As an exercise try to evaluate D using the elements along the second row.
Ax = C (A13)
A.4 INVERSE OF A MATRIX 621
x = A-1C
x1 - x2 + x3 = - 1
-x1 + x2 + x3 = - 1 (A16)
x1 + 2x2 - 2x3 = 5
Here
1 -1 1
A = C -1 1 1S
1 2 -2
*See Kreyszig, E., Advanced Engineering Mathematics, John Wiley & Sons, Inc., New York.
622 APPENDIX A M AT R I X A L G E B R A FOR S T R U C T U R A L A N A LY S I S
-1 -1
` ` - ` ` ` `
1 1 1 1
2 -2 1 -2 1 2
-1 -1
C = F- ` ` ` ` - ` `V
1 1 1 1
2 -2 1 -2 1 2
-1 -1
` ` - ` ` ` `
1 1 1 1
1 1 -1 1 -1 1
-4 0 -2
CT = C - 1 -3 -2 S
-3 -3 0
Since
1 -1 1
A = -1 1 1 = -6
A 1 2 -2
-4 0 -2
1
A-1 = - C -1 -3 -2 S
6
-3 -3 0
x1 -4 0 -2 -1
1
C x2 S = - C - 1 -3 - 2 S C -1 S
6
x3 -3 -3 0 5