0% found this document useful (0 votes)
4 views26 pages

M2 Week 2

The document discusses methods for solving linear systems, including Gaussian elimination and finding the inverse of square matrices. It explains the conditions for unique, no, or infinite solutions based on the determinant of the matrix. Additionally, it covers row reduction techniques and the concepts of echelon form and reduced echelon form in matrix operations.

Uploaded by

trunghuu2404
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
0% found this document useful (0 votes)
4 views26 pages

M2 Week 2

The document discusses methods for solving linear systems, including Gaussian elimination and finding the inverse of square matrices. It explains the conditions for unique, no, or infinite solutions based on the determinant of the matrix. Additionally, it covers row reduction techniques and the concepts of echelon form and reduced echelon form in matrix operations.

Uploaded by

trunghuu2404
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

Finding Solutions of Linear Systems

Gaussian Ellimination

Inverse of a Square Matrix


Linear Equations
 a11 x1 + a12 x2 = b1

a21 x1 + a22 x2 = b2
Ax = b
 a11 a12 
A =   Det A = a11a22 − a12 a21
 a21 a22 
Det A ≠ 0 ⇒ unique solution
Det A = 0 ⇒ no solution or ∞ solutions
 x1 − 2 x2 = −1  1 − 2  x1   − 1
    =  
− x1 + 3x2 = 3  − 1 3  x2   3 
 x1 − 2 x2 = −1  1 −2   x1   −1
    =  
− x1 + 2 x2 =3  −1 2   x2   3 
 x1 − 2 x2 =−1  1 −2   x1   −1
    =  
− x1 + 2 x2 =
1  −1 2   x2   1 
Solution of linear systems
• Lets consider 3 by 3 linear system
 a11 x1 + a12 x2 + a13 x3 =
b1

a21 x1 + a22 x2 + a23 x3 =
b2
a x + a32 x2 + a33 x3 =
b3
 31 1
Solution of linear systems
• Example: 3 by 3 linear system
1 −2 1 2   x −2 y=
+z 2 1 −2 1   x   2 
0 1 −1 0  ⇔  y − z =0 ⇔ 0 1 −1  y  = 0 
       
0 0 2 4   2z = 4 0 0 2   z   4 

z = 2;
y= z= 2;
x = 2 + 2 y − z = 2 + 4 − 2 = 4;
 4
 
x =  2
 2
 
Solution of linear systems
• Example: Find the solution of the linear system
4 x − 2 y + 5 z = 16 4 − 2 5 16 
 1 1 
 x + y = 0  0 0 
 x + 3 y − 2 z = −6 1 3 − 2 − 6
  
1 1 0 0  1 1 0 0 
1 3 −2 −6  R2 → R2 − R1  
  R → R − 4R  0 2 −2 −6 
3 3 1
 4 −2 5 16  0 −6 5 16 
1 1 0 0 
R2 → R2 / 2 0 1 −1 −3
R3 → R3 + 6 R2  
0 −6 5 16 
1 1 0 0  − z =−2 y =−3 + z =−3 + 2 =−1
0 1 −1 = − 3  z 2 = x+ y 0
 
0 0 −1 −2  y − z =−3 x =− y =1
Solving Linear Systems
• Augmented matrix
 a11 a12  a1n b1 
a a22  a2 n b2  Ax = b
 21
     
 
am1 am 2  amn bm 
• Elementary row operations
1. Add to one row a multiple of another row
2. Interchange two rows
3. Multiply all elements of a row by a nonzero constant

• A nonzero row/column is a row/column which contains at least one


nonzero entry
• A leading entry of a row refers to the leftmost nonzero entry

• A leading entry of a column refers to the uppermost nonzero entry


Row Reduction and Echelon Form
• A rectangular matrix is in echelon form (EF) if it has the following
properties
1. All nonzero rows are above any rows of all zeros
2. Each leading entry of a row is in a column to the right of the leading
entry of the row above it
3. All entries in a column below a leading entry are zeros

• A matrix in echelon form with the following additional conditions is


called reduced echelon form (REF) if
• − nonzero number
1. The leading entry in each non zero row is 1 * − any number
2. Each leading 1 is the only nonzero entry in its column

2 − 4 − 6 1  1 0 0 − 7   • * * *
0 1  0 1 0 4  0 • * * 
 8 − 3    
0 0 0 4  0 0 1 3  0 0 0 0
Row Reduction and Echelon Form

• * * * * * * * 1 * * 0 0 * 0 *
0 0 0 • * * * * 0 0 0 1 0 *
  0 *
0 0 0 0 • * * * 0 0 0 0 1 * 0 *
   
0 0 0 0 0 0 • 0 0 0 0 0 0 0 1 0
• Echelon form Reduced echelon form REF or (RREF)

• A pivot position in a matrix A is a location in A that corresponds to


the leading 1 in the reduced echelon form

• The dots identify the pivots positions

• Pivots are leading elements in a row for a matrix in echelon form


Row Reduction
An Example: Reduce given matrix into REF form
• Step 1: Start from the leftmost nonzero column,
The pivot position is at the top

0 3 − 6 6 4 − 5
pivot column 3 − 7 8 − 5 8 9 
 
3 − 9 12 − 9 6 15 

• Step 2: Select a nonzero entry in the pivot column as a pivot. If


necessary interchange rows to move this entry into the pivot position
pivot
3 − 9 12 − 9 6 15 
3 − 7 8 − 5 8 9  R1 ↔ R3
 
0 3 − 6 6 4 − 5
Row Reduction
An Example:
• Step 3: Use row operations to create zeroes in all positions below
the pivot
3 − 9 12 − 9 6 15  3 − 9 12 − 9 6 15 
3 − 7 8 − 5 8 9  R2 → R2 − R1 0 2 − 4 4 2 − 6 
   
0 3 − 6 6 4 − 5
0 3 − 6 6 4 − 5
• Step 4: Cover the row containing the pivot position and cover all rows
if any above it. Apply the Steps 1-3 to the remaining matrix.
• Repeat the process until there are no more nonzero rows to modify

3 − 9 12 − 9 6 15 
• Pivot 0 2 − 4 4 2 − 6
 
0 3 − 6 6 4 − 5
Pivot column: c2
Row Reduction
An Example:
• Step 4: Use row operations to create zeroes in all positions below
the pivot
3 − 9 12 − 9 6 15  3 − 9 12 − 9 6 15 
0 2 − 4 4 2 − 6  R → R / 2 0 1 − 2 2 1 − 3
  2 2
 
0 3 − 6 6 4 − 5 0 3 − 6 6 4 − 5

3 − 9 12 − 9 6 15 
0 1 − 2 2 1
 − 3 R3 → R3 − 3R2
0 3 − 6 6 4 − 5
3 − 9 12 − 9 6 15 
0 1 − 2 2 1
 − 3 This is in echelon form
0 0 0 0 1 4 
Reduced Echelon Form
• Step 5: Start with the rightmost pivot create zeros above each
pivot
3 − 9 12 − 9 6 15  R2 → R2 − R3 3 − 9 12 − 9 0 − 9 
0 1 − 2 2 1 − 3 0 1 − 2 2 0 − 7 
  R1 → R1 − 6R3  
0 0 0 0 1 4  0 0 0 0 1 4 

3 − 9 12 − 9 0 − 9  3 0 − 6 9 0 − 72
0 1 − 2 2 0 − 7  R1 → R1 + 9R2 0 1 − 2 2 0 − 7 
   
0 0 0 0 1 4  0 0 0 0 1 4 

1 0 − 2 3 0 − 24
0 1 − 2 2 0 − 7  R1 → R1 / 3 This is reduced echelon form
  • Steps 1-4 are called the
0 0 0 0 1 4  forward phase;
• Step 5 is called backward
phase
Row Reduction
• An Example: Find EF and RREF form of the matrix
 0 −3 −6 4 9  1 4 5 − 9 − 7
 −1 − 2 −1 3 1  R1 ↔ R4  −1 − 2 −1 3 1 
   
− 2 − 3 0 3 − 1 − 2 − 3 0 3 − 1
   
1 4 5 − 9 − 7  0 −3 −6 4 9 

• All elements below the pivot must be zero R2 → R2 + R1


R3 → R3 + 2R1
R3 → R3 − 5 2 R2
R → R +3 2R R3 ↔ R4
4 4 2

1 4 5 − 9 − 7  1 4 5 − 9 − 7  1 4 5 − 9 − 7 
0 2 4 − 6 − 6    0 2 4 − 6 − 6 
   0 2 4 − 6 − 6  
0 5 10 − 15 − 15 0 0 0 0 0  0 0 0 − 5 0 
    
 0 − 3 − 6 4 9   0 0 0 − 5 0  0 0 0 0 0 
Row Reduction
• An Example 1 4 5 − 9 − 7   • * * * *
0 2 4 − 6 − 6  0 • * * * 
   
0 0 0 − 5 0  0 0 0 • * 
   
 0 0 0 0 0  0 0 0 0 0 
• A rectangular matrix is in echelon form if it has the following
properties
1. All nonzero rows are above any rows of all zeros
2. Each leading entry of a row is in a column to the right of the leading
entry of the row above it
3. All entries in a column below a leading entry are zeros

• A matrix in echelon form with the following additional conditions is


called reduced echelon form if

1. The leading entry in each non zero row is 1


2. Each leading 1 is the only nonzero entry in its column
Row Reduction
• An Example: Find EF and RREF form of the matrix
1 4 5 − 9 − 7  1 4 5 −9 −7 
0 2 4 − 6 − 6 R3 → R3 / (−5) 0 2 4 −6 −6 
  
0 0 0 − 5 0  0 0 0 1 0
   
 0 0 0 0 0  0 0 0 0 0
• All elements above pivots must be zero
1 4 5 0 −7  1 0 −3 0 5
R2 → R2 + 6 R3 0 2 4 0 −6  0 2 4 0 −6 
  R1 → R1 − 2 R2 
R1 → R1 + 9 R3 0 0 0 1 0  0 0 0 1 0
   
0 0 0 0 0  0 0 0 0 0
1 0 −3 0 5 
0 1 2 0 −3
 
R2 → R2 / 2 0 0 0 1 0 
 
0 0 0 0 0 
Solution of linear systems
• Example: An augmented matrix reduced into echelon form (EF) .
Find all solutions of corresponding linear system
1 1 − 5 1 
0 1 1 4 
 
0 0 0 0
• Associate linear system has the form
 x1 + x2 − 5 x3 = 1 
 x + x = 4 
 2 3 
 0 = 0
• Variables x1 and x2 : basic or leading variables; x3 is a free variable
• To write the solution express basic variables x1 and x2 in terms of
free variable x3.
x2 = 4 − x3 , x1 = 1 − x2 + 5 x3
x1 = 1 − x2 + 5 x3 = 1 − (4 − x3 ) + 5 x3 = −3 + 6 x3
Solution of linear systems
• Example: An augmented matrix reduced into reduced echelon form
(REF) . Find all solutions of corresponding linear system
1 0 − 5 1 
0 1 1 4 
 
0 0 0 0
• Associate linear system has the form
 x1 − 5 x3 = 1 
 x + x = 4 
 2 3 
 0 = 0
• Variables x1 and x2 corresponding the pivot columns are called the
basic or leading variables; x3 is called a free variable
• To write the solution express basic variables x1 and x2 in terms of
free variable x3. x1 = 1 + 5 x3 , x2 = 4 − x3
x3 is free
Solution of linear systems
• Example: For which value of h the system is inconsistent ?
 x + 5 y − 3 z = −4  1 5 − 3 − 4
  −1 − 4 1 3 
 − x − 4 y + z = 3  
− 2 − 7 y =h − 2 − 7 0 h 

1 5 −3 −4 
R2 → R2 + R1 0
 1 − 2 − 1 
R3 → R3 + 2 R1
0 3 − 6 h − 8
1 5 −3 −4 
0
 1 − 2 − 1  h ≠ 5
R3 → R3 − 3R2
0 0 0 h − 5
1 5 − 3 −4 
0 1 − 2  0 = h−5 ≠ 0 Inconsistency, no solution
 − 1 
0 0 0 h − 5 ≠ 0
Solution of linear systems
• Example: Find all solutions of linear system for h=5

1 5 −3 −4 
0 1 −2 −1  h = 5
R3 → R3 − 3R2  
0 0 0 h − 5
1 5 −3 −4  z − is free var x + 5 y − 3z =−4
0 1 −2 −1 y − 2 z =−1 x =−4 − 5 y + 3 z
 
0 0 0 0  y =−1 + 2 z x =−4 − 5(−1 + 2 z ) + 3 z
x, y − are basic var x =−4 + 5 − 10 z + 3 z =1 − 7 z

 x   1− 7z   1   − 7z   1   − 7
           
 y  =  − 1 + 2 z  =  − 1 +  2 z  =  − 1  + z  2 
z  z   0   z   0   1 
           
Solution of linear systems
• Example:

 x   1− 7z   1   − 7z   1   − 7
           
 y  =  − 1 + 2 z  =  − 1 +  2 z  =  − 1 + z  2 
z  z   0   z   0   1 
           
z =t∈R
 x   1  − 7  x 1 − 7
           
 y  =  − 1 + t  2 , r =  y  a =  − 1, v =  2 
z  0   1  z 0  1 
           
r = a + vt

Equation of a line though the point a along the direction v


For 3 x 3 systems

x1 −2 x2 +2 x3 =1
x1 −2 x2 + x3 =1
3 x1 −2 x2 + x3 =1

Put the equations in matrix form:

1 −2 2   x1  1 1 − 2 2
1 −2 1   x  = 1 1 − 2 1 
  2    
3 −2 1   x3  1 3 − 2 1 

If we can find an inverse of this matrix, we can solve this system:


Finding the inverse of a 3 x 3 matrix

The inverse can be found by using a series of row operations:

1. Multiplying a row by a scalar

2. Adding or subtracting two rows.

3. Swapping any two rows

x1 −2 x2 +2 x3 =1
x1 −2 x2 + x3 =1
3 x1 −2 x2 + x3 =1
To find the inverse:
1 − 2 2
1 − 2 1 
1. First write the matrix of the linear system:  
3 − 2 1 

1 − 2 2 1 0 0
2. “Augment” the Identity matrix: 1 − 2 1 0 1 0
 
3 − 2 1 0 0 1

3. Use row operations to transform the left half into the identity matrix I.
The right half will then be the inverse.
Example: Solve the system of equations x1 − 2 x2 + 2 x3 = 1
x1 − 2 x2 + x3 = 1
Augmented matrix: 3 x1 − 2 x2 + x3 = 1

1 −2 2 1 0 0 1 −2 2 1 0 0
1 −2 1 R2 → R2 − R1 R2 ↔ R3
 0 1 0  ⇔ 0 0 −1 −1 1 0 
R3 → R3 − 3R1
3 −2 1 0 0 1  0 4 −5 −3 0 1 
1 −2 2 1 0 0 1 −2 2 1 0 0
0 4 −5 R2 → R2 + 5 R3
 −3 0 1  R3 → (−1) × R3 0 4 −5
 −3 0 1 
R1 → R1 − 2 R3
0 0 −1 −1 1 0  0 0 1 1 −1 0 
1 −2 0 −1 2 0  2 0 0 0 −1 1  R1 → R1 / 2
0 4 0 2 −5 1  R1 → 2 R1 − R2 0 4 0 2 −5 1  R2 → R2 / 4
 
0 0 1 1 −1 0   0 0 1 1 −1 0 
1 0 0 0 −1/ 2 1/ 2   0 −1/ 2 1/ 2 
0 1 0 1/ 2 −5 / 4 1/ 4  ⇒ A −=
1 1/ 2 −5 / 4 1/ 4 
  
0 0 1 1 −1 0   1 −1 0 
So the solution to
x1 − 2 x2 + 2 x3 = 1 1 −2 2 
A 1 −2 1 
= A −1A = I
x1 − 2 x2 + x3 = 1
3 −2 1 
3 x1 − 2 x2 + x3 = 1
 0 −1/ 2 1/ 2 
A −1 1/ 2 −5 / 4 1/ 4 
=
 1 −1 0 

is:
 0 −1/ 2 1/ 2  1  0 
x= 1/ 2 −5 / 4 1/ 4  1 =
A −1b =  −1/ 2 
    
 1 −1 0  1  0 
Example: Find the inverse of a matrix A by a series of row-operations:

1 1 0 1 0 0  1 1 0 
 2 2 1 0 1 0
  A =  2 2 1 
1 2 3 0 0 1 1 2 3 
1 1 0 1 0 0
0 0 1 − 2 1 0 
R2 → R2 – 2 R1  
R3 → R3 – R1 0 1 3 − 1 0 1

1 1 0 1 0 0
 
R2 → R3 0 1 3 − 1 0 1
R3 → R2 0 0 1 − 2 1 0

1 1 0 1 0 0  −4 3 −1 
0 1 0 5 − 3 1 
R2 → R2 – 3 R3   A −1  5 −3 1 
=
0 0 1 − 2 1 0  −2 1 0 

R1 → R1 – R2 1 0 0 − 4 3 − 1
0 1 0 5 − 3 1 
  A −1A = I
0 0 1 − 2 1 0 

You might also like