Scilab/C4/LinearequationsIterativeMethods/English
Title of script: Solving System of Linear Equations using Iterative Methods
Author: Shamika
Keywords: System of linear equations, Iterative Methods



Slide 1  Dear friends, welcome to the Spoken Tutorial on “Solving System of Linear Equations using Iterative Methods” 
Slide 2 Learning Objective Slide  At the end of this tutorial, you will learn how to:

Slide 3System Requirement slide  To record this tutorial, I am using

Slide 4 Prerequisites slide  Before practising this tutorial, a learner should have basic knowledge of

Slide 5 Jacobi Method  * The first iterative method we will be studying is Jacobi method.

Slide 6 Example  Let us solve this example using Jacobi Method 
Switch to Scilab and open JacobiIteration_final.sci  Let us look at the code for Jacobi Method. 
Highlight
format('e',20) 
We use format method to specify the format of the displayed answers on the Scilab console.

Highlight
A=input("Enter the coeffiecient matrix of nxn: ") b=input("Enter the righthand side matrix nx1: ") x0=input("Initial approximation nx1: ") MaxIter=input("Maximum no. of iterations: ") tol=input("Enter the convergence tolerance :")//stop if norm change in x < tol 
Then we use input function to get the values for

Highlight
[m, n] = size( A ) 
Then we use size function to check if A matrix is a square matrix.

Highlight
for i=1:1:m sum=0; for j=1:1:m sum=sum+abs(A(i,j)); end if 2*abs(A(i,i))<sum then error("the matrix is not diagonally dominant") end end 

Highlight
function [ solution ] = JacobiIteration( A, b, x0, MaxIter, tol ) 
Then we define the function Jacobi Iteration with input arguments A, b , x zero, maximum iteration and tolerance level.

Highlight
if ( length(x0) ~= n ) error( "Sizes of input matrix and input vector do not match" ) end 
We check if the size of A matrix and initial values matrix are compatible with each other. 
Highlight
xk = x0 for k = 1 : 1 : MaxIter for i = 1 : n xkp1( i ) = (b(i)  A(i, 1:i1)*xk(1:i1)  A(i, i+1:n)*xk(i+1:n)) / A(i,i) //Computes the jacobian updates end
printf( "iter = %d, Relative Error = %g\n", k, RelativeError ) xk = xkp1
break end end


Click on Execute and select Save and Execute  Let us save and execute the function. 
Switch to Scilab console  Switch to Scilab console. 
Enter these values on Scilab console for each prompt
Enter the coeffiecient matrix of nxn: [2 1;5 7] 
Let us enter the values at each prompt.
The coefficient matrix A is open square bracket two space one semi colon five space seven close square bracket

Type
Enter the righthand side matrix nx1: [11; 13] 
Then we type open square bracket eleven semicolon thirteen Press Enter. 
Type
Initial approximation nx1: [1;1] 
* The initial values matrix is open square bracket one semi colon one close square bracket
Press Enter. 
Type
Maximum no. of iterations: 25 
The maximum number of iterations is twenty five.
Press Enter. 
Type
Enter the convergence tolerance :0.00001

Let the convergence tolerance level be zero point zero zero zero zero one
Press Enter. 
Type:
JacobiIteration( A, b, x0, MaxIter, tol ) 
We call the function by typing
Jacobi Iteration open paranthesis A comma b comma x zero comma M a x I t e r comma t o l close paranthesis Press Enter.

Slide 7 Gauss Seidel Method  Let us now study Gauss Seidel method.

Slide 8 Gauss Seidel  In Jacobi method, for the computation of x of i k plus one, every element of x of i k is used except x of i k plus one

Slide 9 Example  Let us solve this example using Gauss Seidel Method 
Switch to Scilab editor and open GaussSeidel.sci  Let us look at the code for Gauss Seidel Method 
Highlight
format('e',20) 
The first line specifies the format of the displayed answer on the console using format function. 
Highlight
A=input("Enter the coeffiecient matrix of nxn: ") b=input("Enter the righthand side matrix nx1: ") x0=input("Initial approximation nx1: ") MaxIter=input("Maximum no. of iterations: ") tol=input("Enter the convergence tolerance :")//stop if norm change in x < tol 
Then we use input function to get the values of

Highlight
function [ solution ] = GaussSeidel( A, b, x0, MaxIter, tol ) 
Then we define the function Gauss Seidel with input arguments A comma b comma x zero comma max iterations and tolerance level and output argument solution

Highlight
// Check that the matrix is square [m, n] = size( A ) if ( m ~= n ) error( "The input matrix is not square" ) end
if ( length(x0) ~= n ) error( "Sizes of input matrix and input vector do not match" ) end 
We check if matrix A is square and the sizes of initial vector and matrix A are compatible using size and length function. 
Highlight
xk = x0 xkp1 = zeros( xk ) 
Then we start the iterations.

Highlight
for k = 1 : 1 : MaxIter
for i = 1 : n xkp1( i ) = (b(i)  A(i, 1:i1)*xkp1(1:i1)  A(i,i+1:n)*xk(i+1:n)) / A(i,i) // x^{k+1} end // Applies the relaxation RelativeError = norm( xkp1  xk, 'inf' ) / norm( xkp1, 'inf' ) printf( "iter = %d, Relative Error = %g\n", k, RelativeError ) xk = xkp1 if ( RelativeError < tol ) break end end

We solve for each equation to get the value of the unknown variable for that equation using x k p one.

Click on Execute and select Save and Execute  Let us save and execute the function. 
Switch to Scilab console  Switch to Scilab console 
Type the following on Scilab console:
For first prompt [2 1;5 7] 
For the first prompt, we type matrix A.
Type open square bracket two space one semi colon five space seven close square bracket Press Enter.

Second prompt
[11; 13] 
For the next prompt,
type left square bracket eleven semi colon thirteen right square bracket Press Enter. 
For third prompt
[1;1] 
We provide the values of initial value vector by typing
open square bracket one semicolon one close square bracket

For fourth prompt
25 
Then we specify the maximum number of iterations to be twenty five

For fifth prompt
0.00001 
Let us define tolerance level to be zero point zero zero zero zero one

Then type:
GaussSeidel( A, b, x0, MaxIter, tol )

* Finally we call the function by typing
G a u s s S e i d e l open paranthesis A comma b comma x zero comma M a x I t e r comma t o l close paranthesis

Slide 13 Solve  Solve this problem on your own using Jacobi and Gauss Seidel methods 
Slide 14 Summary  In this tutorial, we have learnt to:

Show Slide 15
Title: About the Spoken Tutorial Project

* About the Spoken Tutorial Project

Show Slide 16
Title: Spoken Tutorial Workshops The Spoken Tutorial Project Team

The Spoken Tutorial Project Team

Show Slide 17
Title: Acknowledgement

* Spoken Tutorial Project is a part of the Talk to a Teacher project

This is Ashwini Patil signing off. Thank you for joining.
