ME 261: Numerical Analysis

Class Problems #5 - Solution of System of Linear Equations

______________________________________________________________________________
1. (i) Solve the following system of linear equations ([A]{x}={b}) using Naïve Gauss-
Elimination method:
x1 + x2 − x3 = 2
2 x1 + 3x2 − 5x3 = −3
3x1 + 2 x2 − 3x3 = 6
(ii) Using the results obtained (i), find the [L] matrix required for solving the system with
LU decomposition method.
(iii) Using the results obtained in (i), find the determinant of the coefficient matrix [A]
(iv) Repeat (i), (ii) and (iii) using Gauss-elimination method with partial pivoting.
(v) Repeat (i), (ii) and (iii) using Gauss-elimination method with scaled-partial pivoting.

2. (i) Solve the following system of linear equations ([A]{x}={b}) using Naïve Gauss-
Elimination method and keep at least four digits after decimal in your calculations:
x1 + x2 − 2 x3 = 3
4 x1 − 2 x2 + x3 = 5
3x1 − x2 + 3x3 = 8
(ii) Using the results obtained (i), find the [L] matrix required for solving the system with
LU decomposition method.
(iii) Using the results obtained in (i), find the determinant of the coefficient matrix [A]
(iv) Repeat (i), (ii) and (iii) using Gauss-elimination method with partial pivoting.
(v) Repeat (i), (ii) and (iii) using Gauss-elimination method with scaled-partial pivoting.

3. (i) Solve the following system of linear equations ([A]{x}={b}) using a suitable
elimination (Naïve Gauss-Elimination) method:
3x2 + 2 x3 − 2 x4 = 4
x1 + x2 + x3 + 5x4 = −2
2 x1 + x2 − 2 x3 = −1
3x1 + x2 − x3 + 3x4 = 4

1|Page
 (ii) Using the results obtained (i), find the [L] matrix required for solving the system with
LU decomposition method.
(iii) Using the results obtained in (i), find the determinant of the coefficient matrix [A]
(iv) Repeat (i), (ii) and (iii) using Gauss-elimination method with partial pivoting.
(v) Repeat (i), (ii) and (iii) using Gauss-elimination method with scaled-partial pivoting.

4. (i) Solve the following system of linear equations ([A]{x}={b}) using a suitable
elimination (Gauss Jordan without partial pivoting) method and keep at least four digits
after decimal in your calculations:
3x2 + 2 x3 − 2 x4 = 4
x1 + x2 + x3 + 5x4 = −2
2 x1 + x2 − 2 x3 = −1
3x1 + x2 − x3 + 3x4 = 4
(ii) Repeat (i) using Gauss Jordan with partial pivoting.
(iii) Repeat (i) using Gauss Jordan with scaled-partial pivoting.

5. Find the inverse of the following matrix in order to solve a system of linear equations in
the form [A]{x} = {b} using (i) Gauss-Elimination method, (ii) Gauss Jordan method, (iii)
LU decomposition method:
2 0 1
A = 3 2 5
1 −1 0

6. There are two systems of linear equations [ A]{ X } = {bi } ; i = 1, 2

With
1 2 −1 0 2  −3
3    −3
4 0 1 151  
[ A] =  ; {b1} =   ; {b2 } =  
0 2 5 4   35  0
   30   3 
1 2 3 4
Solve the above two systems using LU decomposition method.

2|Page
 7. Use L-U factorization to solve the system Ax=B
where,
2 1 5  5
   
A = 4 4 −4 , B = 0
1 3 1   
6
With partial pivoting. [Four decimal places accuracy required]

8. Find the solution of the following system of equations using LU decomposition technique.
x1 + 3x2 + 3x3 = 16
2 x1 + x2 + 4 x3 = 16
3x1 + 2 x2 + x3 = 10

9. (i) Use LU decomposition method to invert Matrix [A] given below:

 4 1 6
[ A] = 1 3 1
5 2 5

(ii) Use the inverse to solve

20
 
[ A]{ X } = {b}, where{b} = 10 
24
 

10. Idealized spring-mass systems have numerous applications throughout engineering. An

arrangement of four springs in series being depressed with a force of 2000 kg. At
equilibrium, force-balance equations can be developed defining the interrelationships
between the springs,
k2 ( x2 − x1 ) = k1 x1
k3 ( x3 − x2 ) = k2 ( x2 − x1 )
k4 ( x4 − x3 ) = k3 ( x3 − x2 )
F = k4 ( x4 − x3 )
where the k’s are spring constants k1 through k4 are 150, 50, 75, and 225 N/m, respectively.
(i) Solve the above problem using Gauss-Elimination method.

3|Page
 (i) Suppose now this problem needs to be solved using LU decomposition method. Find
the [L] and [U] matrices. Also, show how the solution vector {x} can be found using [L]
and [U] matrix (only show the matrix equations).
(iii) Can you solve this problem using iterative methods? Explain the justification of using
SOR (successive over-relaxation) to solve this problem.

11. Solve the following system of linear equations by Jacobi’s method:

 6 −2 1   x1  11
 1 2 −5  x  = −1
  2  
−2 7 2   x3   5 

Assume any suitable initial guess. Consider || Eabsolute ||  0.1 . (Note E stands for error

between two consecutive iteration).

12. Consider the following linear system,

 4 1 −1 1  x1  0
      (0)  
Ax = b , A =  −2 5 0  , b = −7 , x =  x2  , xIniitialVector = 1
2 1 6 13  x  1
     3  
(i) Solve the above system of equations using the gauss-Seidel method. Show at least one
iteration.
(ii) Will this system converge to a solution using the Gauss-Seidel method?

13. Solve the following linear system of equations by the Gauss-Seidel method. Show at least
five iterations:
 3 −1 0 −1  x1  −3
−1 3 −1 0   x   2 
   2  =  
 0 −1 4 −1  x3   6 
 
−1 0 −1 4   x4  12 

Start with the initial value of x0 = 0 0 0 0

T

14. (i) Solve the following system of equations by the Gauss-Seidel method:

4|Page
 2 x1 − 7 x2 − 10 x3 = −17
5x1 + x2 + 3x3 = 14
x1 + 10 x2 + 9 x3 = 7

Use x1(0) = x2(0) = x3(0) = 0 as initial guesses. Perform 10 iterations maintaining at least four

digits after decimal for all the iterations. x1(0) = x2(0) = x3(0) = 0

(ii) Did the iterations in the previous problem (i) converge? Explain your answer by
mentioning the general criterion for a system of equations to converge in case of the Gauss-
Seidel method and the associated prevailing situation in the system of equations above.

15. Solve the following system of linear equations correct to three significant figures by Gauss-
Seidel method
 45 2 3   x1  58
−3 22 2   x  = 47
  2  
 5 1 20  x3  67 

Take initial values of unknowns as (0,0,0)T .

16. Solve the following system of equations using Gauss-Seidel method.

 6 −2 1   x1  11
 1 2 −5  x  = −1
  2  
−2 7 2   x3   5 

Show solution for first five iterations.

17. Apply Gauss Seidel method to the following systems for solution:
 3 1 −1  x1  4
 2 4 1   x  = 1
  2  
−1 2 5   x3  1

[Minimum four decimal places accuracy required]

18. In solving a particular problem, the following coefficient matrix has been obtained:

5|Page
 2 + 0.1x2 −1 0 0 0 0 0 0 
 
 −1 2 + 0.1x 2 −1 0 0 0 0 0 
 0 −1 2 + 0.1x −1
2
0 0 0 0 
 
 0 
 
 
 −1 2 + 0.1x2 −1 
 0 0 0 0 0 −1 2 + 0.1x2 −1 
 
 0 0 0 0 0 0 −1 2 + 0.1x2 
(i) Mention whether the direct or iterative approach is recommended for solving the above
system.
(ii) If iterative approach is to be used, determine the required condition for the above system
to converge.
(iii) Formulate the general convergence criterion for the solution of a system of equations
with iterative approach.

19. Determine the roots of the following non-linear system of equations using Gauss-Seidel
and Newton-Raphson methods:
x2 + y 2 = 4
ex + y = 1
Use initial guesses as [x(0), y(0)] = [1, −1.7]. Perform two sets of iterations and comment on
the results.

20. Determine the roots of the following non-linear system of equations using Newton-
Raphson method:
xyz − x2 + y 2 = 1.34
xy − z 2 = 0.09
e x − e y + z = 0.41
Assume any suitable initial guesses and perform one iteration.

6|Page