Name

1. Nada Sayed Abdel Ghaffar

2. Tasneem Abdel Hadi
Subject
Numerical Analysis

Faculty of Education

Major
‫رياضيات عام‬

Code of subject
MATB365
1) Use the Newton- Raphson to find a root of
the equation x =−cosx use an initial guess of x =¿
3
0

1
With ε =0.005 % ? s

Solution:
f ( x )=x 3 +cosx
2
f ( x ) =3 x −sinx

f (x) 3
x + cosx
x i+1=x i− → x i+1=x i− 2
f (x ) 3 x −sinx

N
0 1
1 27.260
1
2 18.174
1
3 12.119
1
4 8.0813
5 5.3752
6 3.5926
7 2.4314
8 1.6344
9 1.0212
10 0.3237
11 263.48
40
12 176.65
62
2) Use the Newton- Raphson to find a root of
the equation
x −x
f ( x )=e + 2 +2 cosx =6 use an initial guess of x 0=1 with ε s=0.005 % ?

Solution: -
f ( x )=e x + 2−x +2 cosx−6=0
x −x
f ( x ) =e −2 −2 sinx=0

f (x) x −x
e +2 + 2 cosx−6
x i+1=x i− → x i+1=x i− x −x
f (x ) e −2 −2 sinx

N
0 1
1 3.4698
2 2.7261
3 2.1973
4 1.9143
ε a=|1.82948−1.8294
1.82948 |× 100 %=0.004< ε =0.005 %
s

5 1.8341
6 1.8294
7 1.8294
8
3) Solving using Newton- Raphson method
to locate the root of f ( x )=sin √ x−x use an initial
guess of x =4 with ε =0.01 % ?
0 s

Solution: -
f ( x )=sin √ x−x

cos √ x
f (x )= −1
2 √x

f (x) sin √ x−x

x i+1=x i− → x i+1=x i−
f (x ) cos √ x
−1
2√x

N ε a= |0.76865−0.7687
0.76865 |× 100 %=0.006 % <ε =0.01 %
s

0 1
1 0.7828
2 0.7687
3 0.7687
4 0.7686
5
4) Use the fixed-point iteration method to
estimate one of the roots of x =2 x+3 with an 2

accuracy of ε =0.05 % at x=4 ? s

Solution: -
x=√ 2 x+3
1
φ ( x )= √ 2 x+ 3φ ( x )=
√ 2 x+3

|φ ( x )|=|√ |1
8+3
=0.3015< 1 success

N φ(x)

0 3.3166
1 3.1037
2 3.0344
3 3.0114
ε a= |3.0004−3.0013
3.0004 |× 100 %=0.02 %< ε =0.05 %
s

4 3.0038
5 3.0013
6 3.0004

: use Gauss- seidel method to solve the

Q1

following system:
 [ ][ ] [ ]
2 4 10 −1 x 1 −11
10 −1 2 0 x2 = 6
0 −3 1 −8 x 3 −15
−1 12 −1 3 x 25
4

Assume zero initial guesses and take ε s=0.05 % ?

Solution: - the matrix in question is not diagonally

dominant
Then we make it diagonally dominant by
Re- order it

[ ][ ] [ ]
10 −1 2 0 x1 6
−1 12 −1 3 x 2 = 25
2 4 10 −1 x 3 −11
0 −3 1 −8 x −15
4

Then the matrix is diagonally dominant

1
10 x 1−x 2+2 x 3=6 →
10
[ 6+ x 2−2 x 3 ]
1
−x 1+ 12 x 2−x 3 +3 x 4=25 →
12
[ 25+ x1 + x 3−3 x 4 ]
1
2 x1 + 4 x 2+ 10 x 3−x 4 =−11→
10
[−11−2 x 1−4 x2 + x 4 ]
1
−3 x 2+ x 3−8 x 4=−15 →
−8
[−15+3 x 2−x 3 ]

N x 1 x2 x3 x4

0 0.6 2.133 - 0.815

3 2.0733 8
1 1.2271 1.808 - 0.948
9 1.9874 2
 2 1.1784 1.778 - 0.963
9 1.9524 9
3 1.1684 1.777 - 0.965
0 1.9480 1
4 1.1673 1.777 - 0.965
1.9478 2
5 1.1672 1.776 - 0.965
6 1 1.9474 5
6 1.1670 1.776 - 0.965
9 1.9476 2

|1.1670−1.16726
ε a=
1.1670 |×100 %=0.02 %
ε =| |×100 %=0.04 %
1.7769−1.7761
a
1.7769

ε =| |× 100 %=0.01 %
−1.9476+1.9474
a
−1.9476

ε =| |×100 %=0.03 %
0.9652−0.9655
a
0.9652

: Use Gauss- seidel method to solve the

Q2

following system of equations:

50 x 1+2 x 2−31 x3 +6 x 4 =190
 x 1−3 x 2+2 x 3−55 x 4=60

3 x 1+51 x 2+ 2 x 3 + x 4=80

2 x1 +24 x 2 +33 x 3+ 3 x 4=60

Assume zero initial guesses and take ε s=0.01 % ?

Solution: - this equation is not diagonally

dominant
Then we make it diagonally dominant by
Re- order these equations:
1
50 x 1+2 x 2−31 x3 +6 x 4 =19 0 →
50
[ 190−2 x 2 +31 x3 −6 x 4 ]

1
3 x 1+51 x 2+ 2 x 3 + x 4=8 0→
51
[ 80−3 x 1−2 x 3−x 4 ]
1
2 x1 +24 x 2 +33 x 3+ 3 x 4=6 0 →
33
[ 60−2 x 1−24 x 2−3 x 4 ]

1
x 1−3 x 2+2 x 3−55 x 4=6 0 →
55
[ 60−x 1+3 x 2−2 x 3 ]

x1 x2 x3 x4
N
0 3.8 1.3451 0.6096 -1.0730
1 4.2529 1.3156 1.5782 -1.0271
2 4.8491 1.2416 0.7147 -1.0445
3 4.3188 1.3070 0.7008 -1.0582
4 4.3092 1.3084 0.7017 -1.0584
5 4.3097 1.3083 0.7016 -1.05831
5 8
 6 4.3097 1.3083 0.7016 -1.05840
0 5 7
7 4.3097 1.3083 0.7016 -
09 52 77 1.058391

|4.309709−4.30970
ε a=
4.309709 |×100 %=0.0002 %
ε =
a [ 1.308352−1.30835
1.308352 ] ×100 %=0.0001 %
ε =| |×100 %=0.0009 %
0.701677−0.70167
a
0.701677

ε =| |× 100 %=0.0008 %
−1.058391+1.05840
a
−1.058391

Solution for question 3: -

The matrix is diagonally dominant
Then;
1
4 T 11 −T 12−T 21=175 →
4
[ 175+T 12+ T 21 ]
1
−T 11 +4 T 12−T 22=125 →
4
[ 125+T 11 +T 22]
 1
−T 11 +4 T 21−T 22=75 →
4
[ 75+T 11 +T 22]
1
−T 12−T 21 +4 T 22=25 →
4
[25+ T 12 +T 21 ]

N T 11 T 12 T 21 T 22

0 43.75 42.187 29.687 24.218

5 5 8
1 61.718 52.734 40.234 29.492
8 4 4 2
2 66.992 55.371 42.871 30.810
2 1 1 5
3 68.310 56.030 43.530 31.222
5 3 3 5
4 68.640 56.195 43.695 31.222
1 1 1 53
5 68.722 56.236 43.736 31.243
5 3 3 1
6 68.743 56.246 43.746 31.248
1 6 6 3
7 68.748 56.249 43.749 31.249
3 1 1 6
8 68.749 56.249 43.749 31.249
6 8 8 9
9 68.749 56.249 43.749 31.249
9 9 9 1

|68.7499−68.7496
ε a=
68.7499 |× 100 %=0.0004 %
ε =| |×100 %=0.0001 %
56.2499−56.2498
a
56.2499
 |
ε a=
43.7499−43.7498
43.7499 |×100 %=0.0002 %

ε =
a [ 31.2491−31.2499
31.2491 ] ×100 %=0.002 %

Try solve
1) Solving using Newton- Raphson method to
locate the root of f ( x )=sin √ x−x use an initial
guess of x =0.5 withε =0.01 % ?
0 s

Solution : -
f ( x )=sin √ x−x

cos √ x
f (x )= −1
2 √x

f (x) sin √ x−x

x i+1=x i− → x i+1=x i−
f (x ) cos √ x
−1
2√x
N

0 0.5
ε a=|0.768949−0.76865
0.768949 |× 100 %=0.0001 %
1 0.8236

2 0.7697

3 0.76865

4 0.768949

2) Use Gauss- seidel method to solve the

following system of equations:
−3 x 1+ x 2 +12 x3 =50
6 x 1−x 2−x 3=3
 6 x 1+ 9 x2 + x 3=40

Assume zero initial guesses and take ε s=0.5 % ?

Solution: - these equations aren’t diagonally

dominant
Then we make it diagonally dominant by
Re- order these equations:
1
6 x 1−x 2−x 3=3 →
6
[ 3+ x 2+ x 3 ]
1
6 x 1+ 9 x2 + x 3=40 →
9
[ 40−6 x 1−x 3 ]
1
−3 x 1+ x 2 +12 x3 =50→ [ 50+3 x 1−x 2 ]
12
x1 x2 x3
N
0 0.5 4.111 3.949
1 1
1 1.843 2.754 4.264
4 3 1
2 1.669 2.867 4.284 ε a=
[ 1.6905 ]
1.6905−1.6919
×100 %=0.08 %
7 1 1
3 1.691 2.857 4.287
9 1 1
4 1.690 2.857 4.287
5 1 1
|2.8571−2.8671
ε a=
2.8571 |×100 %=0.4 %
ε =| |×100 %=0.07 %
4.2871−4.2841
a
4.2871