Escolar Documentos
Profissional Documentos
Cultura Documentos
1.2 Adjoint :
Let A = [aij] be a square matrix of order n. Let Aij be the cofactor of aij. Then the nth order matrix [Aij]T is called the adjoint of A. It is denoted by adjA. Thus the adjA is nothing but the transpose of the cofactor matrix [Aij] of A. Result : If A is a square matrix of order n, then A (adjA) = | A | In = (adj A) A, where In is the identity matrix of order n. Proof : Let us prove this result for a square matrix A of order 3. a11 a12 a13
2 1
2 3 =3 1 3 1 3 = 9 2 3 1 2 =5 2 1 1 1 =4 1 3 1 1 =1 2 3 1 1 =3 2 1 1 1 =5 2 3
1 1 =4 1 3 1 1 = =1 1 2
9 5 1 4 4 1 3
3 1 2 Example 1.3 : If A = , verify the result A (adj A) = (adj A) A = | A | I2 1 4 1 2 1 2 A= Solution: , | A | = =2 1 4 1 4 4 2 adj A = 1 1 1 2 4 2 2 0 1 0 (1) A (adj A) = = =2 = 2I2 0 1 1 4 1 1 0 2 4 2 1 2 2 0 1 0 (adj A) A = (2) = = 2 = 2I2 0 1 1 1 1 4 0 2 From (1) and (2) we get A (adj A) = (adj A) A = | A | I2.
1 5 4 1
Example 1.4 : If A = 1 2
1 adj A = 9
1 2 1
3
1 4
1 3
| A | = 1 2 1 A (adj A) = 1 2
1
1 2 1 1 2
(adj A) A = 9 5
= 11 0 1 0 = 11 I3 = | A | I3 0 0 1 1 1 1 3 4 5 1 3 1 0 0
1 1 0 0
= 1(6 3) 1 (3 + 6) + 1( 1 4) = 11 3 1 11 0 0 3 4 5 3 9 1 4 = 0 11 0 0 0 11 3 5 3 1
1 3 (1) 0 11 0
4 1 1 2
2 1
11 3 = 0 3 0
11
0 0 (2)
1.3 Inverse :
Let A be a square matrix of order n. Then a matrix B, if it exists, such that AB = BA = In is called inverse of the matrix A. In this case, we say that A is an invertible matrix. If a matrix A possesses an inverse, then it must be unique. To see this, assume that B and C are two inverses of A, then AB = BA = In (1) AC = CA = In Now AB = In C(AB) = CIn (CA)B = C InB = C B = C (2) (Q associative property)
i.e., The inverse of a matrix is unique. Next, let us find a formula for computing the inverse of a matrix. We have already seen that, if A is a square matrix of order n, then A(adj A) = (adj A)A = | A | In
If we assume that A is non-singular, then | A | 0. Dividing the above equation by | A |, we get 1 1 A | A | (adj A) = | A | (adj A) A = In. From this equation it is clear that the inverse of A is nothing but 1 1 | A | (adj A). We denote this by A . Thus we have the following formula for computing the inverse of a matrix through its adjoint. If A is a non-singular matrix, there exists an inverse which is given by 1 A1 = | A | (adj A).
1.3.1 Properties :
1. Reversal Law for Inverses : If A, B are any two non-singular matrices of the same order, then AB is also non-singular and (AB)1 = B1 A1 i.e., the inverse of a product is the product of the inverses taken in the reverse order. Proof : Since A and B are non-singular, | A | 0 and | B | 0. We know that | AB | = | A | | B | | A | 0, | B | 0 | A | | B | 0 | AB | 0 Hence AB is also non-singular. So AB is invertible. (AB) (B1A1) = A (BB1)A1 = AIA1 = AA1 = I Similarly we can show that (B1A1) (AB) = I (AB) (B1A1) = (B1A1) (AB) = I B1 A1is the inverse of AB. (AB)1 = B1 A1 2. Reversal Law for Transposes (without proof) : If A and B are matrices conformable to multiplication, then (AB)T = BTAT.
i.e., the transpose of the product is the product of the transposes taken in the reverse order. 3. For any non-singular matrix A, (AT)
1
= (A 1)
Proof : We know that AA1 = I = A1 A Taking transpose on both sides of AA1 = I, we have (AA1) = IT By reversal law for transposes we get (A1) AT = I Similarly, by taking transposes on both sides of A A = I, we have AT(A1) = I From (1) & (2)
T T (A1) AT = AT (A1) = I T 1 T T
(1) (2)
= (A 1)
Solution:
2 1 1 1 4 2 = | A | (adj A) = 2 = 1 1 1 1 2 2
. 1
0
Then | A | =
2 1
2 2
=20 1
0
A is non-singular and hence A 1 exists 2 0 Cofactor of 3 = A11 = =2 2 1 2 0 Cofactor of 1 = A12 = = 2 1 1 2 2 Cofactor of 1 = A13 = =6 1 2 1 1 Cofactor of 2 = A21 = =1 2 1 3 1 Cofactor of 2 = A22 = = 2 1 1 3 1 Cofactor of 0 = A23 = = 5 1 2
1 1 =2 2 0 3 1 = =2 2 0 3 1 = =8 2 2
[Aij] = 1 2
2
2 1 2 2 5 ; adj A = 2 2 2 6 5 8 2 8 2 1 2 1 1 A1 = | A | (adj A) = 2 2 2 2 6 5 8
2 6
= 1 3
Solution: | A | = 1 0 and | B | = 1 0 So A and B are invertible.
1 1 2 1 1 5 2
1 4
adj B =
2 1 1 0
(1) (2)
2 1 1 2 1 3 = 1 0 1 1 1 2
From (1) and (2) we have (AB)1 = B1 A1. EXERCISE 1.1 (1) Find the adjoint of the following matrices : 1 2 3 2 5 3 3 1 0 5 0 (ii) (i) (iii) 3 1 2 2 4 2 4 3 1 2 1 1 2 (2) Find the adjoint of the matrix A = and verify the result 3 5 A (adj A) = (adj A)A = | A | . I 3 3 4 (3) Find the adjoint of the matrix A = 2 3 4 and verify the result 1 A (adj A) = (adj A)A = | A | . I (4) Find the inverse of each of the following matrices : 1 0 3 1 3 7
(i) 2 1 1 1 1 1 8 1 3 2 (iv) 5 1 10 1 4
(ii) 4 2 3 1 2 1
1 (iii) 1 0
2 3 2
2 0
2 2 1 (v) 1 3 1 1 2 2
(5) If A =
(7)
3 3 4 Find the inverse of the matrix A = 2 3 4 and verify that A3 = A 1 0 1 1 1 2 2 T 1 2 is 3A . Show that the adjoint of A = 2 2 2 1
(10) For A =
10
a22 an2
a
a2n
nn
x b = x b
x1
2
b1
2
(1) where b1
2
a22 an2
n1
x b ; X= ;B= x b a
a2n
nn n n
If the coefficients matrix A is non-singular, then A1 exists. Pre-multiply both sides of (1) by A1 we get A1 (AX) = A1B (A1A)X = A1B IX = A1B X = A1B is the solution of (1) Thus to determine the solution vector X we must compute A1. Note that this solution is unique. Example 1.7 : Solve by matrix inversion method x + y = 3, 2x + 3y = 8 Solution: The given system of equations can be written in the form of 1 1 x 3 = 2 3 y 8 AX = B 1 1 |A| = =10 2 3
Here
11
x 3 1 3 = y 2 1 8 x 1 = y 2
x = 1, y = 2 Example 1.8 : Solve by matrix inversion method 2x y + 3z = 9, x + y + z = 6, xy+z=2 Solution : The matrix equation is 2 1 3 x
1 1 1 1 2 A X = B, where A = 1 1
1 1
9 y = 6 z 2 9 x 1 1, X = y and B = 6 z 2 1 1 2 1 3 | A | = 1 1 1 = 2 0 1 1 1
1 3
A is a non-singular matrix and hence A1 exists. The cofactors are A11 = 2, A12 = 0, A13 = 2 A21 = 2, A22 = 1, A23 = 1, A31 = 4, A32 = + 1, A33 = 3 The matrix formed by the cofactors is 2 0
[Aij] = 2 4
2 1
1 1
12
2 The adjoint of A = 0 2
2 4 1 1
1 1 Inverse of A = | A | (adj A) 2 2 4 1 1 0 1 1 A = 2 2 1 3
= adj A 3
The solution is given by X = A1B 2 2 4 9 x y = 1 0 1 1 6 2 z 2 1 3 2 2 1 1 4 = 2 = 2 3 6 x = 1, y = 2, z = 3 EXERCISE 1.2 Solve by matrix inversion method each of the following system of linear equations : (1) 2x y = 7, 3x 2y = 11 (2) 7x + 3y = 1, 2x + y = 0 (3) x + y + z = 9, 2x + 5y + 7z = 52, 2x + y z = 0 (4) 2x y + z = 7, 3x + y 5z = 13, x+y+z=5 (5) x 3y 8z + 10 = 0, 3x + y = 4, 2x + 5y + 6z = 13
Definition :
The matrix A is said to be of rank r, if (i) A has atleast one minor of order r which does not vanish. (ii) Every minor of A of order (r + 1) and higher order vanishes. In other words, the rank of a matrix is the order of any highest order non vanishing minor of the matrix. The rank of A is denoted by the symbol (A). The rank of a null matrix is defined to be zero. The rank of the unit matrix of order n is n. The rank of an m n matrix A cannot exceed the minimum of m and n. i.e., (A) min {m, n}. 7 1 Example 1.9 : Find the rank of the matrix 2 1 7 1 Solution : Let A = . This is a second order matrix. 2 1 The highest order of minor of A is also 2. 7 1 The minor is given by =90 2 1 The highest order of non-vanishing minor of A is 2. Hence (A) = 2. 2 4 Example 1.10 : Find the rank of the matrix 1 2 2 4 Solution : Let A = . 1 2 2 4 The highest order minor of A is given by = 0. Since the second 1 2 order minor vanishes (A) 2. We have to try for atleast one non-zero first order minor, i.e., atleast one non-zero element of A. This is possible because A has non-zero elements (A) = 1. 1 2 3 Example 1.11 : Find the rank of the matrix
2 5
4 1
1
6
Solution : Let A = 2
5
1
2 4 1
1
3 6 14
2 5
4 1
1 6 = 2 1 5 1
2 2 1
=0 1
3 3
Since the third order minor vanishes, (A) 3 2 4 = 22 0 5 1 A has atleast one non-zero minor of order 2. (A) = 2 1 1 1 3 Example 1.12 : Find the rank of the matrix
2 5
1 3 1 7
11
4
Solution : Let A = 2 5
1
1 3
1 7 This is a matrix of order 3 4 A has minors of highest order 3. They are given by 1 1 1 1 1 3
11
3 4
2 5
1 3
1 1 1
2 3 5 7
= 0 ; 2 1 4 = 0 ; 5 1 11 7 1 1 3 3 4 = 0 ; 1 3 4 = 0 1 7 11 11
All the third order minors vanish. (A) 3 Next, we have to try for atleast one non-zero minor of order 2. This is 1 1 possible, because A has a 2nd order minor = 3 0 (A) = 2 2 1 Note : In the above examples, we have seen that the determination of the rank of a matrix involves the computation of determinants. The computation of determinants may be greatly reduced by means of certain elementary transformations of its rows and columns. These transformations will greatly facilitate our dealings with the problem of the determination of the rank and other allied problems.
15
16
In the following examples we illustrate the method of finding the rank of matrices by reducing them to the echelon form. 1 1 1 Example 1.13 : Find the rank of the matrix
2 3
3 2
3
4
Solution : Let A = 2 3
4 3 2 3 1 1 1 0 5 6 0 5 6 1 1 1 0 5 6 0 0 0
1 1
R2 R2 2R1 R3 R3 3R1
R3 R3 R2
The last equivalent matrix is in echelon form. The number of non-zero rows is 2. (A) = 2 1 2 3 1 Example 1.14 : Find the rank of the matrix
2 3
4 6 2 6 9
1 Solution : Let A = 2 3
2 3 1 4 6 2
6 9 1 2 3 1 0 0 0 0 R2 R2 2R1 R3 R3 3R1 0 0 0 0
This equivalent matrix is in the echelon form. Since the number of non-zero rows of the matrix in this echelon form is 1, (A) = 1. 4 2 1 3 Example 1.15 : Find the rank of the matrix 6 3 4 7 2 1 0 1
17
4 2 1 3 Solution : Let A = 6 3 4 7 2 1 0 1 1 2 4 3 4 3 6 7 C C 3 1 0 1 2 1 4 3 1 2 0 5 10 5 R R 4R 2 1 2 0 1 2 1 1 2 4 3 0 1 2 1 R 1 R 5 2 2 0 1 2 1 1 2 4 3 0 1 2 1 R R R 3 2 3 0 0 0 0
The last equivalent matrix is in the echelon form. The number of non-zero rows in this matrix is two. (A) = 2 3 1 5 1 Example 1.16 : Find the rank of the matrix 3 1 5 1
Solution : Let A = 1 2
1 5 1 5 1 5 7 2 5 7 1 2 1 5 R2 R2 3R1 0 7 8 14 R3 R3 R1 0 7 8 7 1 2 1 5 0 7 8 14 R3 R3 R2 0 0 0 7
1 1 1 3 1
5 7 2 1
2 5 1 2
R1 R2
The last equivalent matrix is in the echelon form. It has three non-zero rows. (A) = 3
18
(4)
1 3 2 0 2 1
1 2 3 1 3 1
3 4
2 0
6 12 6 (2) 1 2 1 4 8 4
3 1 2 0 (3) 1 0 1 0 2 1 3 0
3
1 0
1
1 (5) 2 3
2 1 4 6 1 3
2 7
1 (6) 2 1
2 4 2
3 7
1 3
6
4
19
20
b 2 bn
Cramers Rule for Non homogeneous equations of 2 unknowns : Let us start with the system of two linear equations in two unknowns x and y. (i) a11x + a12y = b1 a21x + a22y = b2 a11 a12 Let = a21 a22 x. = x = (ii)
b1 a12y b2 a22y
a12 a22
21
= = x. = y. =
b1 a12 a12 a12 y (by properties of determinants) b2 a22 a22 a22 b1 a12 y . 0 (by properties of determinants) b2 a22 b1 a12 = x (say) b2 a22 a11 b1 = y (say) a21 b2
Similarly
x, y are the determinants which can also be obtained by replacing 1st and 2nd column respectively by the column of constants containing b1 and b2 i.e. by x b1 b Thus, we have, x = x x = 2 y = y y = y provided 0
Since , x, y are unique, there exists a unique solution for the above system of equations. i.e., the system is consistent and has a unique solution. The method stated above to solve the system of equation is known as Cramers Rule. Cramers rule is applicable when 0. If = 0, then the given system may be consistent or inconsistent. Case 1 : If = 0 and x = 0, y = 0 and atleast one of the coefficients a11, a12, a21, a22 is non-zero, then the system is consistent and has infinitely many solutions. Case 2 : If = 0 and atleast one of the values x, y is non-zero, then the system is inconsistent i.e. it has no solution. To illustrate the possibilities that can occur in solving systems of linear equations with two unknowns, consider the following three examples. Solve : (1) x + 2y = 3 x+y=2 (2) x + 2y = 3 2x + 4y = 6 (3) x + 2y = 3 2x + 4y = 8
22
Unique solution
1 2 = = 1 1 1 3 2 = = 1 2 1 1 3 = = 1 1 2
S O
X X+2Y = 3 X+Y = 2
Fig. 1.4 Since 0, the system has unique solution. By Cramers rule x y x= =1 ; y= = 1 (x, y) = (1, 1) Solution (2) : Infinitely many solution Y 1 2 We have = = 0 2 4 3 2 x = = 0 o 6 4 X S 1 3 X + 2Y = 3 y = = 0 2X + 4Y = 6 2 6 Fig. 1.5 Since = 0 and x = 0, y = 0 and atleast one of a11, a12, a21, a22 is non zero, it has infinitely (case 1) many solutions. The above system is reduced to a single equation x + 2y = 3. To solve this equation, assign y = k x = 3 2y = 3 2k The solution is x = 3 2k, y = k ; k R For different value of k we get different solution. In particular (1, 1), ( 1, 2), (5 1) and (8, 2.5) are some solutions for k = 1, 2, 1 and 2.5 respectively Solution (3) : No Solution Y 1 2 = = 0 ; 2 4 3 2 1 3 x = = 4 ; y = =2 8 4 2 8 o X Since = 0 and x 0, y 0 2X + 4Y = 8 X + 2Y = 3 (case 2 : atleast one of the value of x, y, non-zero), the system is Fig. 1.6 inconsistent. i.e. it has no solution.
23
11 = a21 a31
a
a23 , a33
a13
1 x = b2 b3
b a z =
a33
a13 a23
11 y = a21 a31
a
a23 , a33
a13
11 a21 a31
b3
b1 b2
As we discussed earlier for two variables, we give the following rule for testing the consistency of the above system. Case 1 : If 0, then the system is consistent, and has a unique solution. Using Cramers Rule can solve this system. Case 2 : If = 0, we have three important possibilities. Subcase 2(a) : If = 0 and atleast one of the values of x, y and z is non-zero, then the system has no solution i.e. Equations are inconsistent. Subcase 2(b) : If = 0 and x = y = z = 0 and atleast one of the 2 2 minor of is non zero, then the system is consistent and has infinitely many solution. In this case, the system of three equations is reduced to two equations. It can be solved by taking two suitable equations and assigning an arbitrary value to one of the three unknowns and then solve for the other two unknowns. Subcase 2(c) : If = 0 and x = y = z = 0 and all their (2 2) minors are zero but atleast one of the elements of is non zero (aij 0) then the system is consistent and it has infinitely many solution. In this case, system is reduced to a single equation. To solve we can assign arbitrary values to any two variables and can determine the value of third variable. Subcase 2(d) : If = 0, x = y = z = 0, all 2 2 minors of = 0 and atleast one 2 2 minor of x or y or z is non zero then the system is inconsistent.
24
Theorem 1.2 (without proof) : If a non-homogeneous system of linear equations with more number of unknowns than the number of equations is consistent, then it has infinitely many solutions. To illustrate the different possibilities when we solve the above type of system of equations, consider the following examples. (1) 2x + y + z = 5 (2) x + 2y + 3z = 6 x+y+z=4 x+y+z=3 x y + 2z = 1 2x + 3y + 4z = 9 (3) x + 2y + 3z = 6 (4) x + 2y + 3z = 6 2x + 4y + 6z = 12 x+y+z=3 3x + 6y + 9z = 18 2x + 3y + 4z = 10 (5) x + 2y + 3z = 6 2x + 4y + 6z = 12 3x + 6y + 9z = 24 Solution (1) : 2x + y + z = 5 ; x + y + z = 4 ; x y + 2z = 1 Unique solution We have 2 1 1 = 1 1 1 = 3 1 1 2 5 1 1 x = 4 1 1 = 3 1 1 2 Fig. 1.7
2 1 5 2 5 1 1 4 1 = 6 ; z = 1 1 4 = 3 y = 1 1 2 1 1 1 = 3, x = 3, y = 6, z = 3
0, The system has unique solution. By Cramers rule. x 3 y 6 z x = = 3 = 1, y = = 3 = 2, z = =1 The solution is x = 1, y = 2, z = 1 (x, y, z) = (1, 2, 1)
25
Solution (2) : x + 2y + 3z = 6 ; x + y + z = 3 ; 2x + 3y + 4z = 9 1 2 3 6 2 3 1 1 1 = 0 ; x = 3 1 1 = 0 = 2 3 4 9 3 4 1 6 3 1 2 6 1 3 1 = 0 ; y = z = 1 1 3 = 0 2 9 4 2 3 9 Since = 0 and x = y = z = 0 but atleast one of the 2 2 minors of is 1 2 0, the system is consistent (by case 2(b)) and has non-zero 1 1 infinitely many solution. The system is reduced to 2 equations. Assigning an arbitrary value to one of unknowns, say z = k, and taking first two equations. We get x + 2y + 3k = 6 Infinitely many solution x+y+k = 3 i.e., x + 2y = 6 3k x+y = 3k
1 2 =1 = 1 1
Fig. 1.8 6 3k 2 x = = 6 3k 6 + 2k = k 3 k 1 1 6 3k y = = 3 k 6 + 3k = 2k 3 1 3 k x k x= = = k 1 y 2k 3 = = 3 2k y= 1 The solution is x = k, y = 3 2k and z = k i.e. (x, y, z) = (k, 3 2k, k). k R Particularly, for k = 1, 2, 3, 4 we get (1, 1, 1), (2, 1, 2), (3, 3, 3), (4, 5, 4) respectively as solution.
26
Solution (3) : x + 2y + 3z = 6 ; 2x + 4y + 6z = 12 ; 3x + 6y + 9z = 18 1 2 3 6 2 3 2 4 6 = 0 ; = 12 4 6 = 0 = x 3 6 9 18 6 9 1 6 3 1 2 6 y = 2 12 6 = 0 ; z = 2 4 12 = 0 3 18 9 3 6 18 Here = 0 and x = y = z = 0. Also all their 2 2 minors are zero, but atleast one of aij of is non- zero. It has infinitely many solution (by case 2(c)). The system given above is reduced to one equation i.e. x + 2y + 3z = 6 Assigning arbitrary values to two of the three unknowns say y = s, z = t We get x = 6 2y 3z = 6 2s 3t The solution is x = 6 2s 3t, y = s, z = t i.e. (x, y, z) = (6 2s 3t, s, t) s, t R For different value s, t we get different solution. Solution (4) : x + 2y + 3z = 6 ; x + y + z = 3 ; 1 2 3 = 1 1 1 = 0 2 3 4 2x + 3y + 4z = 10 No Solution Infinitely many solution
Fig. 1.9
6 2 3 x = 3 1 1 = 1 10 3 4
Fig. 1.10 Since = 0, x 0 (atleast one of the values of x, y, z non-zero) The system is inconsistent (by case 2(a)). It has no solution.
27
Here = 0 and x = y = z = 0. All the 2 2 minors of are zero, but we see that atleast one of the 2 2 minors of x or y or z is non zero. i.e. 12 4 0 minor of 3 in x 24 6 by case 2(d), the system is inconsistent and it has no solution.
Fig. 1.11
Example 1.17 : Solve the following system of linear equations by determinant method. (1) x + y = 3, 2x + 3y = 7 (2) 2x + 3y = 8, 4x + 6y = 16 (3) x y = 2, 3y = 3x 7
=76=1
y 1 = 1 = 1
28
Solution (2) : 2x + 3y = 8 ; 4x + 6y = 16 2 3 = = 12 12 = 0 4 6 8 3 x = = 48 48 = 0 16 6 2 8 y = = 32 32 = 0 4 16 Since = 0, and x = y = 0 and atleast one of the coefficients aij of 0, the system is consistent and has infinitely many solutions. All 2 2 minor are zero and atleast (1 1) minor is non zero. The system is reduced to a single equation. We assign arbitrary value to x (or y) and solve for y (or x). Suppose we assign x = t, from equation (1) 1 we get y = 3 (8 2t). 8 2t The solution set is (x, y) = t, 3 , t R In particular (x, y) = (1, 2) (x, y) = ( 2, 4) 1 (x, y) = 2 , 3 for t = 1 for t = 2 1 for t = 2
Solution (3) : x y = 2 ; 3y = 3x 7 1 1 = = 0, 3 3 2 1 x = 7 3 = 1 Since = 0 and x 0 (atleast one of the values x or y 0) the system is inconsistent. It has no solution. Example 1.18 : Solve the following non-homogeneous equations of three unknowns. (1) x + 2y + z = 7 (2) x + y + 2z = 6 (3) 2x + 2y + z = 5 2x y + 2z = 4 3x + y z = 2 xy+z=1 x + y 2z = 1 4x + 2y + z = 8 3x + y + 2z = 4
29
(4)
x + y + 2z = 4 2x + 2y + 4z = 8 3x + 3y + 6z = 12
(5)
x + y + 2z = 4 2x + 2y + 4z = 8 3x + 3y + 6z = 10 x + y 2z = 1
Solution (1) : x + 2y + z = 7, 1 2 1 =
2x y + 2z = 4,
2 1 2 = 15 1 1 2 7 2 1 x = 4 1 2 = 15 1 1 2 1 2 7 z = 2 1 4 = 30 1 1 1
1 7 1 ; y = 2 4 2 = 30 1 1 2
= 15, x = 15, y = 30, z = 30 By Cramers rule x y z x= = 1, y = = 2, z = = 2 Solution is (x, y, z) = (1, 2, 2) 3x + y z = 2, 4x + 2y + z = 8 Solution (2) : x + y + 2z = 6, 1 1 2 6 1 2
2 8 8 Since = 0 and x = y = z = 0, also atleast one of the (2 2) minors of is not zero, the system is consistent and has infinitely many solution. Take two suitable equations and assign arbitrary value to one of the three unknowns. We solve for the other two unknowns. Let z = k R equation (1) and (2) becomes x + y =6 2k 3x + y = 2 + k
4 1 y = 3 4
= 3 1 1 = 0, 2 6 2
1 2 1 = 0, 1
x = 2 1 1 = 0,
8 1 z = 3 4
2 1 1 6 1 2 = 0
30
= x = y =
1 1 =13=2 3 1 6 2k 1 = 6 2k 2 k = 4 3k 2 + k 1 1 6 2k = 2 + k 18 + 6k = 7k 16 3 2 + k
x 4 3k 1 = = 2 (3k 4) 2
By Cramers rule x= y=
y 7k 16 1 = = 2 (16 7k) 2
Particular Numerical solutions for k = 2 and 2 are ( 5, 15, 2) and (1, 1, 2) respectively Solution (3) : 2x + 2y + z = 5, 2 2 1 x y + z = 1,
= 1 1 1 = 0
5 ; x = 1 1 1 0 4 1 2
3x + y + 2z = 4 2 1
Since = 0 and x 0 (atleast one of the values of x, y, z non zero) the system is inconsistent. i.e. it has no solution. Solution (4) : x + y + 2z = 4, 1 1 2 2x + 2y + 4z = 8, 4 1 2 3x + 3y + 6z = 12
= 2 2 4 = 0 3 3 6 1 4 2 y = 2 8 4 = 0, 3 12 6
x = 8 2 4 = 0 12 3 6 1 1 4 z = 2 2 8 = 0 3 3 12
31
Since = 0 and x = y = z = 0 also all 2 2 minors of , x, y and z are zero, by case 2(c), it is consistent and has infinitely many solutions. ( all 2 2 minors zero and atleast one of aij of 0, the system is reduced to single equation). Let us take x = s and y = t, we get from equation (1) 1 z = 2 (4 s t) the solution set is 4 s t (x, y, z) = s, t, 2 , s, t R Particular numerical solution for (x, y, z) = (1, 1, 1) 3 (x, y, z) = 1, 2, 2 Solution (5) : x + y + 2z = 4, 2x + 2y + 4z = 8, when s = t = 1 when s = 1, t = 2 3x + 3y + 6z = 10
1 1 2 = 2 2 4 = 0 3 3 6 4 1 2 x = 8 2 4 = 0, 10 3 6
1 4 2 y = 2 8 4 = 0 3 10 6 1 1 4 z = 2 2 8 = 0 3 3 10
= 0 and x = y = z = 0. Also all 2 2 minors of = 0, but not all the minors of x, y and z are zero. Therefore the system is inconsistent. i.e. it has no solution. Example 1.19 : A bag contains 3 types of coins namely Re. 1, Rs. 2 and Rs. 5. There are 30 coins amounting to Rs. 100 in total. Find the number of coins in each category. Solution : Let x, y and z be the number of coins respectively in each category Re. 1, Rs. 2 and Rs. 5. From the given information x + y + z = 30 (i) x + 2y + 5z = 100 (ii) Here we have 3 unknowns but 2 equations. We assign arbitrary value k to z and solve for x and y.
32
The solution is (x, y, z) = (3 k 40, 70 4k, k) k R. Since the number of coins is a non-negative integer, k = 0, 1, 2 Morever 3k 40 0, and 70 4k 0 14 k 17 The possible solutions are (2, 14, 14), (5, 10, 15), (8, 6, 16) and (11, 2, 17)
33
As an illustration, consider a homogeneous linear system of two equations in two unknowns. x+y = 0 xy = 0 the graph of these equations are lines through the origin and the trivial solution corresponding to the point of intersection Fig. 1. 12 at the origin.
X Y =0
S o X
X +Y =0
For the following system X+ Y= 0 2X + 2Y = 0 xy = 0 2x 2y = 0 O the graph shows, that the system has X infinitely many solutions. There is one case in which a Fig. 1.13 homogeneous system is assured of having non-trivial solutions, namely, whenever the system involves more number of unknowns than the number of equations. Theorem 1.3 : (without proof) A homogeneous system of linear equations with more number of unknowns than the number of equations has infinitely many solutions. Example 1.20 : Solve : x + y + 2z = 0 2x + y z = 0 2x + 2y + z = 0 Solution : 1 1 2 = 2 1 1 = 3 1 0, the system has unique solution. The above system of homogeneous equation has only trivial solution. i.e., (x, y, z) = (0, 0, 0).
2 2
34
Since = 0, it has infinitely many solutions. Also atleast one 2 2 minors of 0, the system is reduced to 2 equations. Assigning arbitrary value to one of the unknowns, say z = k and taking first and last equations. (Here we can take any two equations) we get x + y = 2k 2x + y = k 2k 1 1 2k 1 1 = y = = 3k, = 5k = 1, x = 2 1 k 1 2 k By Cramers Rule x = 3k, y = 5k Solution is (x, y, z) = (3k, 5k, k) EXERCISE 1.4 Solve the following non-homogeneous system of linear equations by determinant method : (1) 3x + 2y = 5 (2) 2x + 3y = 5 x + 3y = 4 4x + 6y = 12 (3) 4x + 5y = 9 8x + 10y = 18 2x + y z = 4 x + y 2z = 0 3x + 2y 3z = 4 x + 2y + z = 6 3x + 3y z = 3 2x + y 2z = 3 (4) x+y+z=4 xy+z=2 2x + y z = 1 3x + y z = 2 2x y + 2z = 6 2x + y 2z = 2 2x y + z = 2 6x 3y + 3z = 6 4x 2y + 2z = 4
(5)
(6)
(7)
(8)
35
1 2 1 2 4 1 (9) x + y z = 1 ; x + y + z = 5 ;
3 2 2 x yz=0
(10) A small seminar hall can hold 100 chairs. Three different colours (red, blue and green) of chairs are available. The cost of a red chair is Rs.240, cost of a blue chair is Rs.260 and the cost of a green chair is Rs.300. The total cost of chair is Rs.25,000. Find atleast 3 different solution of the number of chairs in each colour to be purchased.
a32 a33 a3n 3 am2 am3 amn m m1 system and it is denoted by [A, B]. The condition for the consistency of a system of simultaneous linear equations can be given interms of the coefficient and augmented matrices.
31
a a a
The system of simultaneous linear equations AX = B is consistent if and only if the matrices A and [A, B] are of the same rank.
36
The solution of a given system of linear equations is not altered by interchanging any two equations or by multiplying any equation by a non-zero scalar or by adding a multiple of one equation to another equation. By applying elementary row operations to the augmented matrix the given system of equations can be reduced to an equivalent system and this reduced form is used to test for consistency and to find the solutions. Steps to be followed for testing consistency : (i) Write down the given system of equations in the form of a matrix equation AX = B. (ii) Find the augmented matrix [A, B] of the system of equations. (iii) Find the rank of A and rank of [A, B] by applying only elementary row operations. Column operations should not be applied. (iv) (a) If the rank of A rank of [A, B] then the system is inconsistent and has no solution. (b) If the rank of A = rank of [A, B] = n, where n is the number of unknowns in the system then A is a non-singular matrix and the system is consistent and it has a unique solution. (c) If the rank of A = rank of [A, B] < n, then also the system is consistent but has an infinite number of solutions. Example 1.22 : Verify whether the given system of equations is consistent. If it is consistent, solve them.
2x + 5y + 7z = 52,
x + y + z = 9,
2x + y z = 0
Solution : The given system of equations is equivalent to the single matrix equation. 2 5 7 x 52
1 1 1 y = 9 2 1 1 z 0
AX = B
1 9 2 1 1 0
52
37
1 2 2 1 0 0
1 5
1 7
1 1 0 1 1 9 R R2 2R1 3 5 34 2 R3 R3 2R1 1 3 18 9 1 1 1
52 R1 R2
0 1 3 18 R R 3 2 0 3 5 34
0 0
1 (A, B) = 3 1 Also A
1 0
1 4
1 3 18
20
9
R3 R3 + 3R2
The last equivalent matrix is in the echelon form. It has three non-zero rows.
0 0
1 0
1 3
4
1
0 0
1 3 0
y = z
18 20
x+y+z = 9 y 3z = 18 4z = 20
38
(3) z = 5 ; (2) y = 18 3z = 3 ; (1) x = 9 y z x = 935 = 1 Solution is x = 1, y = 3, z = 5 Example 1.23 : Examine the consistency of the equations 2x 3y + 7z = 5, 3x + y 3z = 13, 2x + 19y 47z = 32 Solution : The given system of equations can be written in the form of a matrix equation as 2 3 7 x 5
3 2
1 19
47
3 3 1
y = 13 z 32
AX = B
2 [A, B] = 3 2
7 3
19 47 3 1 2
13 32
5 7 2 5 2
1 3 2
2 3
3 2 1
7 2 3
19 47
1 13 R1 2 32
5 2
R1
R 0 11 27 11 R 2 2 2 0 22 54 27 3 7 5 1 2 2 2 0 11 27 11 R 2 2 2 0 0 0 5
R2 3R1 R3 2R1
R3 4R2
The last equivalent matrix is in the echelon form. It has three non-zero rows. [A, B] = 3 and (A) = 2 (A) [A, B] The given system is inconsistent and hence has no solution. Note : This problem can be solved by not dividing R1 by 2 also. i.e., R2 2R2 3R1 Example 1.24 : Show that the equations x + y + z = 6, x + 2y + 3z = 14, x + 4y + 7z = 30 are consistent and solve them.
39
1 2 3 y = 14 1 4 7 z 30
[A, B] = 1 2 3 14 1 4 7 30 1 1 1 6 R R R 2 1 0 1 2 8 2 R3 R3 R1 0 3 6 24 1 1 1 6
0 1 2 8 R R 3R 3 2 3 0 0 0 0
In the last equivalent matrix, there are two non-zero rows. (A, B) = 2 and (A) = 2 (A) = (A, B) The given system is consistent. But the value of the common rank is less than the number of unknowns. The given system has an infinite number of solutions. The given system is equivalent to the matrix equation 1 1 1 x 6
x+y+z = 6 (1) y + 2z = 8 (2) (2) y = 8 2z ; (1) x = 6 y z = 6 (8 2z) z = z 2 Taking z = k, we get x = k 2, y = 8 2k ; k R Putting k = 1, we have one solution as x = 1, y = 6, z = 1. Thus by giving different values for k we get different solutions. Hence the given system has infinite number of solutions. Example 1.25 : Verify whether the given system of equations is consistent. If it is consistent, solve them :
0 1 2 y = 8 0 0 0 z 0
x y + z = 5,
x + y z = 5,
2x 2y + 2z = 10
40
1 2
1 2
y = 5 z 10
AX = B
2 2 1 1 1 5 0 0 0 0 R2 R2 + R1 R3 R3 2R1 0 0 0 0 In the last equivalent matrix, there is only one non-zero row [A, B] = 1 and (A) = 1 Thus (A) = [A, B] = 1. the given system is consistent. Since the common value of the rank is less than the number of unknowns, there are infinitely many solutions. The given system is equivalent to the matrix equation.
1 [A, B] = 1 2
1 1
1 5
10
5
1 1 1 x 5 0 0 0 y = 0 0 0 0 z 0
x y + z = 5 ; Taking y = k1, z = k2, we have x = 5 + k1 k2. for various values of k1 and k2 we have infinitely many solutions. k1, k2 R Example 1.26 : Investigate for what values of , the simultaneous equations x + y + z = 6, x + 2y + 3z = 10, x + 2y + z = have (i) no solution (ii) a unique solution and (iii) an infinite number of solutions. Solution : The matrix equations corresponding to the given system is 1 1 1 x 6
1 2 3 y = 10 1 2 z
AX = B
41
[A, B] = 1 2 3 10
6 R R2 R1 1 2 4 2 R3 R3 R2 0 3 10 Case (i) : 3 = 0 and 10 0 i.e. = 3 and 10. In this case (A) = 2 while [A, B] = 3 (A) [A, B] The given system is inconsistent and has no solution. Case (ii) : 3 0 i.e., 3 and can take any value in R. In this case (A) = 3 and [A, B] = 3 (A) = [A, B] = 3 = number of unknowns. The given system is consistent and has a unique solution. Case (iii) : 3 = 0 and 10 = 0 i.e., = 3 and = 10 In this case (A) = [A, B] = 2 < number of unknowns. The given system is consistent but has an infinite number of solutions.
1 2 1 0 0
1 1
a [A, B] = a
21
a22 a2n
m1
am2 amn
= [A, O] 0
0 42
Since rank of A = rank of [A, O] is always true, we see that the system of homogeneous equations is always consistent. Note that x1 = 0, x2 = 0, x3 = 0 xn = 0 is always a solution of the system. This solution is called a trivial solution. If the rank of A = rank of [A, B] < n then the system has non trivial solutions including trivial solution. If (A) = n then the system has only trivial solution. Example 1.27 : Solve the following homogeneous linear equations. x + 2y 5z = 0, 3x + 4y + 6z = 0, x + y + z = 0 Solution : The given system of equations can be written in the form of matrix equation 1 2 5 x 0
3 4 1 1
6 y = 0 1 z 0 AX = B
5 6 1 2 2 1 2 1 2 2 1 0
0 0 5 21 6 5 6 21 5 6 9
1 0 0 1 0 0 1 0 0
This is in the echelon form.
0 0 0 0 0 0 0
0 0
R2 R2 3R1 R3 R3 R1
R2 R3
R3 R3 2R2
Clearly [A, B] = 3. and. (A) = 3 (A) = [A, B] = 3 = number of unknowns. The given system of equations is consistent and has a unique solution. i.e., trivial solution. x = 0, y = 0 and z = 0
43
Note : Since (A) = 3, | A | 0 i.e. A is non-singular ; The given system has only trivial solution x = 0, y = 0, z = 0 Example 1.28 : For what value of the equations x + y + 3z = 0, 4x + 3y + z = 0, 2x + y + 2z = 0 have a (i) trivial solution, (ii) non-trivial solution. Solution : The system of equations can be written as AX = B 1 1 3 x 0
4 3 2 1 2 1 [A, B] = 4 2
y = 0 z 0 1 3 0 3 0 1 2 0
8 Case (i) : If 8 then 8 0 and hence there are three non-zero rows. [A] = [A, B] = 3 = the number of unknowns. The system has the trivial solution x = 0, y = 0, z = 0 Case (ii) : If = 8 then. [A, B] = 2 and (A) = 2 (A) = [A, B] = 2 < number of unknowns. The given system is equivalent to x + y + 3z = 0 ; y + 4z = 0 y = 4z ; x=z Taking z = k, we get x = k, y = 4k, z = k [k R {0}] which are non-trivial solutions. Thus the system is consistent and has infinitely many non-trivial solutions. Note : In case (ii) the system also has trivial solution. For only non-trivial solutions we removed k = 0.
0 1 0 1 1 1 0 1 0 0
1 1
3 4 3 12
12 0
0 0 0 0
0
R2 R2 4R1 R3 R3 2R1 R3 R3 R2
44
EXERCISE 1.5 (1) Examine the consistency of the following system of equations. If it is consistent then solve the same. (i) 4x + 3y + 6z = 25 x + 5y + 7z = 13 2x + 9y + z = 1 (ii) x 3y 8z = 10 3x + y 4z = 0 2x + 5y + 6z 13 = 0 (iii) x + y + z = 7 x + 2y + 3z = 18 y + 2z = 6 (iv) x 4y + 7z = 14 3x + 8y 2z = 13 7x 8y + 26z = 5 (v) x + y z = 1 2x + 2y 2z = 2 3x 3y + 3z = 3 (2) Discuss the solutions of the system of equations for all values of . x + y + z = 2, 2x + y 2z = 2, x + y + 4z = 2 (3) For what values of k, the system of equations kx + y + z = 1, x + ky + z = 1, x + y + kz = 1 have (i) unique solution (ii) more than one solution (iii) no solution
45