SOLVING SYSTEM CE 316:

OF LINEAR Numerical
Solutions to
CE Problems

EQUATIONS

Fernando, JPP
 LINEAR SYSTEMS

A System of Linear Equations is when we have two or

more linear equations working together.

𝑨𝒙 = 𝑩
NOTE:
▪ No. of Columns = No. of Unknowns
▪ No. of Rows = No. of Equations

Fernando, JPP
UNIQUE AND MULTIPLE Solving
System of
Linear
SOLUTIONS Equations

Fernando, JPP
 LINEAR SYSTEMS

1. Unique Solution
2. No Solution
3. Infinitely Many Solutions

Fernando, JPP
 LINEAR SYSTEMS

1. Unique Solution

Each column of the right side of the augmented

matrix has a leading one (1).

1 0 0 𝑎
0 1 0 𝑏൩
0 0 1 𝑐

Fernando, JPP
 LINEAR SYSTEMS

2. No Solution

The left of the augmented matrix has a zero row

and the right of it a number not equal to zero.

1 0 0 𝑎
0 1 0 𝑏൩
0 0 0 𝑐

Fernando, JPP
 LINEAR SYSTEMS

2. Infinitely Many Solutions

If we have a zero row.

1 0 0 𝑎
0 1 0 𝑏൩
0 0 0 0

Fernando, JPP
 TRIVIAL AND NON- Solving
System of
Linear
TRIVIAL SOLUTIONS Equations

Fernando, JPP
 TRIVIAL AND NON-TRIVIAL SOLUTIONS
(HOMOGENOUS SYSTEMS)
A system of linear equations is said to be homogenous if it is of
the form:
𝑎 11 𝑥 1 + 𝑎 12 𝑥 2 + ⋯ + 𝑎 𝑛 𝑥 𝑛 = 0
𝑎 21 𝑥 1 + 𝑎 22 𝑥 2 + ⋯ + 𝑎 2𝑛 𝑥 𝑛 = 0
…
𝑎 𝑚1 𝑥 1 + 𝑎 𝑚2 𝑥 2 + ⋯ + 𝑎 𝑚𝑛 𝑥 𝑛 = 0

Which can also be written as:

𝑎 11 𝑎 12 … 𝑎 1𝑛 𝑥1 0
𝑎 21 𝑎 22 … 𝑎 2𝑛 𝑥2 0
… … … … … =
…
𝑎 𝑚1 𝑎 𝑚2 … 𝑎 𝑚𝑛 𝑥𝑛 0

Fernando, JPP
 TRIVIAL AND NON-TRIVIAL SOLUTIONS
(HOMOGENOUS SYSTEMS)
A homogenous system always has at least one
solution.

Consider:
5𝑥 + 4𝑦 = 0
2𝑥 − 2𝑦 = 0

Trivial = Solution set contains zero elements. (0, 0, 0)

Non-Trivial = Solution set contains non-zero elements.

Fernando, JPP
 Solving
System of
DIRECT METHODS Linear
Equations

Fernando, JPP
 DIRECT METHODS

1. Using Inverse of a Matrix

2. Using Gauss Elimination Method

Fernando, JPP
 DIRECT METHODS

Example 1:
𝑥 1 − 2𝑥 2 + 𝑥 3 = 0
6𝑥 2 − 3𝑥 3 = 0
𝑥 1 − 2𝑥 2 − 𝑥 3 = 0

Example 2:
𝑥 1 + 2𝑥 2 + 3𝑥 3 + 2𝑥 4 = 0
𝑥 1 + 3𝑥 2 + 5𝑥 3 + 5𝑥 4 = 0
2𝑥 1 + 4𝑥 2 + 7𝑥 3 + 𝑥 4 = 0
−𝑥1 − 2𝑥 2 − 6𝑥 3 + 7𝑥 4 = 0

Fernando, JPP
GAUSS ELIMINATION Solving
System of
Linear
METHODS Equations

Fernando, JPP
 GAUSS ELIMINATION METHODS

Example 1:

𝑥 1 + 2𝑥 2 + 𝑥 3 = 9
𝑥 1 + 3𝑥 2 + 4𝑥 3 = 11
−6𝑥 2 − 10𝑥 3 = −24

Fernando, JPP
GAUSS ELIMINATION Solving
System of
Linear
WITH ROW PIVOTING Equations

Fernando, JPP
 GAUSS ELIMINATION WITH ROW
PIVOTING
Example 1:

𝑥1 − 𝑥 2 + 𝑥 3 = 2
−6𝑥1 + 𝑥 2 − 𝑥 3 = 3
3𝑥 1 + 𝑥 2 + 𝑥 3 = 4

Fernando, JPP
 Solving
System of
ITERATIVE METHODS Linear
Equations

Fernando, JPP
 ITERATIVE METHODS

Iterative Methods are used because Direct Numerical

Methods are not ideal for very large system of linear
equations and in non-linear equations, Direct
Numerical Methods don’t work.

Fernando, JPP
 Solving
System of
JACOBI METHOD Linear
Equations

Fernando, JPP
 JACOBI METHOD

Given n equations and n variables rearrange each

equation to give a different variable and put in form
x = Tx + c.

Use:
𝑎 1 𝑥1 + 𝑎 2 𝑥 2 + 𝑎 3 𝑥 3 = 𝑎 4
𝑎 5 𝑥1 + 𝑎 6 𝑥 2 + 𝑎 7 𝑥 3 = 𝑎 8
𝑎 9 𝑥1 + 𝑎 10 𝑥 2 + 𝑎 11 𝑥 3 = 𝑎 12

NOTE: This is only for diagonally dominant system of linear

equations!

Fernando, JPP
 JACOBI METHOD

Example:

10𝑥 1 + 2𝑥 2 − 𝑥 3 = 7
𝑥 1 + 8𝑥 2 + 3𝑥 3 = −4
−2𝑥 1 − 𝑥 2 + 10𝑥 3 = 9

Fernando, JPP
 Solving
System of
GAUSS-SEIDEL METHOD Linear
Equations

Fernando, JPP
 GAUSS-SEIDEL METHOD

Gauss-Seidel Method is an improved method of the

Jacobi Method.

𝑥 1 = (−𝑎 2 𝑥 2 − 𝑎 3 𝑥 3 + 𝑎 4 )/𝑎 1
𝑥 2 = (−𝑎 5 𝑥 1 − 𝑎 7 𝑥 3 + 𝑎 8 )/𝑎 6
𝑥 3 = (−𝑎 9 𝑥 1 − 𝑎 10 𝑥 2 + 𝑎 12 )/𝑎 11

Workings for 𝑥 1,2,3 are put in as soon as calculated so

workings to answer quicker.

NOTE: This is only for diagonally dominant system of

linear equations!

Fernando, JPP
 GAUSS-SEIDEL METHOD

Example:

10𝑥 1 + 2𝑥 2 − 𝑥 3 = 7
𝑥 1 + 8𝑥 2 + 3𝑥 3 = −4
−2𝑥 1 − 𝑥 2 + 10𝑥 3 = 9

Fernando, JPP
 CE 316:

END OF PRESENTATION Nu merical

Solutions to CE
Problems

Fernando, JPP