Numerical Analysis Project 2: Systems of Linear Equations

Contributors

• Fatema Moharam ID.6655
• Nourhan Waleed ID.6609
• Heba Elwazzan ID.6521

In this project, we are required to implement 4 numerical methods used to find the roots of a system of linear equations:

• Gaussian-elimination
• LU decomposition
• Gaussian-Jordan
• Gauss-Seidel

Pseudocodes

Gaussian-Elimination

1. Start

2. Input the Augmented Coefficients Matrix (A):

    For i = 1 to n

        For j = 1 to n+1

            Read Ai,j

        Next j

    Next i

3. Apply Gauss Elimination on Matrix A:

    For i = 1 to n-1

        If Ai,i = 0

            Print "Mathematical Error!"
            Stop

        End If

        For j = i+1 to n

            Ratio = Aj,i/Ai,i

            For k = 1 to n+1

                Aj,k = Aj,k - Ratio * Ai,k

            Next k
        Next j
    Next i

4. Obtaining Solution by Back Substitution:

    Xn = An,n+1/An,n

    For i = n-1 to 1 (Step: -1)

        Xi = Ai,n+1

        For j = i+1 to n

            Xi = Xi - Ai,j * Xj

        Next j

        Xi = Xi/Ai,i
    Next i

5. Display Solution:

    For i = 1 to n

        Print Xi

    Next i

6. Stop

LU Decomposition

Gaussian-Jordan

1. Start

2. Input the Augmented Coefficients Matrix (A):

    For i = 1 to n

        For j = 1 to n+1

            Read Ai,j

        Next j

    Next i

3. Apply Gauss Jordan Elimination on Matrix A:

    For i = 1 to n

        If Ai,i = 0

            Print "Mathematical Error!"
            Stop

        End If

        For j = 1 to n

            If i ≠ j 

                Ratio = Aj,i/Ai,i

                For k = 1 to n+1

                    Aj,k = Aj,k - Ratio * Ai,k

                Next k

            End If

        Next j
    Next i

4. Obtaining Solution:

    For i = 1 to n 
        Xi = Ai,n+1/Ai,i
    Next i

5. Display Solution:

    For i = 1 to n

        Print Xi

    Next i

6. Stop

Gauss-Seidel

1. Start

2. Arrange given system of linear equations in 
   diagonally dominant form

3. Read tolerable error (e)

4. Convert the first equation in terms of first variable, 
second equation in terms of second variable and so on. 

5. Set initial guesses for x0,  y0, z0 and so on

6. Substitute value of y0, z0 ... from step 5 in 
   first equation obtained from step 4 to calculate 
   new value of x1. Use x1, z0, u0 .... in second equation 
   obtained from step 4 to caluclate new value of y1. 
   Similarly, use x1, y1, u0... to find new z1 and so on.  

7. If| x0 - x1| > e and | y0 - y1| > e and | z0 - z1|  > e 
   and so on then goto step 9

8. Set x0=x1, y0=y1, z0=z1 and so on and goto step 6

9. Print value of x1, y1, z1 and so on

10. Stop

Analysis of Performance

LU decomposition seems to be the fastest method, but shows more error. Gauss-Seidel is the slowest, as it takes 17 iterations to reach desired accuracy. Gaussian-elimination and Gaussian-Jordan are comparable, with Gaussian-Jordan being the faster of the two.

Sample Runs

Gaussian-elimination

2*a - 6*b - c + 38
-3*a - b + 7*c + 34 
-8*a + b - 2*c + 20

LU decomposition

8*a + 4*b - c - 11
-2*a + 3*b + c - 4 
2*a - b + 6*c - 7

Gaussian-Jordan

2*a + 3*b + c + 4
4*a + b + 4*c - 9 
3*a + 4*b + 6*c

Gauss-Seidel

12*a + 23*b + 5*c - 1
a + 5*b + 3*c - 28
3*a + 7*b + 13*c - 76
1 0 1