MA8491 Numerical Methods 01 - by LearnEngineering - in

For More Visit : www.LearnEngineering.
in
1MA8491 Numerical Methods .
UNIT I : Solution of Equations and Eigenvalue Problems

Solution of equation – Fixed point iteration: x = g(x) method – Newton’s
method – Solution of linear system by Gaussian elimination and Gauss
–Jordon methods – Iterative methods – Gauss-Seidel methods – Inverse of a
matrix by Gauss Jordon method – Eigen value of a matrix by power method
and by Jacobi method for symmetric matrix.
Topic 1 : Solution of equation

A value satisfies given equation is called solution of that equation.
.in
Y axis [ = f(x) ]
–X
x = x1 x = x2 x = x3 x = 0 ing x = x4 x = x5
X axis
eer
gin
–Y
En
Here the curve f(x) with f(x = x2) = f(x = x3) = f(x = x4) = 0
 x2, x3 and x4 are solutions of f(x).
and f(x) > 0 for x2 < x < x3 and x4 < x < x5
arn
f(x) < 0 for x1 < x < x2 and x3 < x < x4
Topic 2 :Fixed point iteration: x = g(x) method

Le
Write the given equation as f(x) = 0.

Find x1, x2 with opposite signs of f(x).
Find an relation as x = g(x) such that g '  x1   1
w.
g '  x2   1
ww
 x , f  x1   f  x2   f  x1  is nearer to zero.
X0   1
 x2 , f  x1   f  x2   f  x2  is nearer to zero.
X1  g  X 0 
X 2  g  X1 
⋮
X n  g  X n 1   Stop method if the consecutive values of
 X n & X n 1 are equal upto required place of decimal.
X n 1  g  X n   i.e., X =X
n n 1
Iteration method is a indirect (or successive approximation) method.
Iteration method is a self correction method.
For More Visit : www.LearnEngineering.in

2 UNIT I : Solution of Equations and Eigenvalue Problems
Part A:
1. If g(x) is continuous in [a,b] then under what condition the iterative
method x = g(x) has unique solution in [a,b].
Sol: g '  x   1 in  a, b  with order of convergence is one.
Part B:
3 2
Eg.1. Solve the equation x  x  100  0 by iteration method.
Sol: Let f  x   x  x  100  0
3 2
f  x  0    0    0   100  100  0  ' ' ve 

3 2
f  x  1  1  1  100  98  0  ' ' ve 

3 2
f  x  2    2    2   100  88  0  ' ' ve 

3 2
.in
f  x  3   3   3  100  64  0  ' ' ve 
3 2
f  x  4    4    4   100  20  0  ' ' ve 

3 2
 f  x  has a root between 4 and 5.

2
ing
f  x  5    5    5   100  50  0  ' ' ve 
eer
i.e, f  x  has a positive root.
gin
This equation f  x   x  x  100  0 can be written as

3 2
x3  x 2  100  0
En
i.e., x 2  x  1  100
10
i.e., x    g  x  , say 
arn
x 1
1
10
 g  x   10  x  1 2
Le
x 1
d
Therefore, g   x    g  x  
dx 
w.
1  1 3
d   1  1 5
 10  x  1 2  10    x  1 2  5  x  1 2 
ww
dx  2 3
  
 x  1 2
5 5
 g  x   3
 3
 x  1 2  x  1 2
5 5 5 1 1
g   x  4       0.57735 1
At x = 4,  4  1 3/ 2
 5 3/ 2
5 5 5 2.23606
5 5
g   x  5    0.34020 1
At x = 5,  5  1 3/ 2
 6 3 / 2

 g   x   1 in (4,5).
So the fixed point iteration method can be applied.
Let x0  4 ∵ f  x  4   f  x  5  
 
Use xn 1  g  xn 
Now, put n = 0 in xn 1  g  xn  , we get
10 10 10
x1  g  x0      4.47214
x0  1 4 1 5 (round off at 5th decimal)
10
x2  g  x1    4.27486
4.47214  1
.in
10
x3  g  x2    4.35406
4.27486  1
ing
10
x4  g  x3    4.32174
4.35406  1
eer
10
x5  g  x4    4.33484
4.32174  1
10
x6  g  x5  
gin
 4.32952
4.33484  1
10
x7  g  x6    4.33168
En
4.32952  1
10
x8  g  x7    4.33080
arn
4.33168  1
10
x9  g  x8    4.33116
4.33080  1
Le
10
x10  g  x9    4.33101
4.33116  1
w.
10
x11  g  x10    4.33107
4.33101  1
ww
10 
x12  g  x11    4.33105  these values are same 
4.33107  1  
 at fifth decimal, i.e., x12  x .
13 
10
x13  g  x12    4.33105  so stop the method 

4.33105  1 
 3 2
The root of the given equation x  x  100  0 is 4.33105
Note: In above problem, the root is 4.33105(at 5th decimal)

or 4.3311(at 4th decimal) or 4.331(at 3rd decimal)

3 2
CW.2. Solve the equation x  2 x  10 x  20 by fixed point iteration
method. {Solution : x13  x14  1.36880 }
3
HW.3. Solve the equationx  x  1  0 by fixed point iteration method.
{Solution : g  x    1  x  ; x23  x24  0.682 }
1/3
Eg.4. Solve the equation 3 x  cos x  2  0 by fixed point iteration

method correct to 3 decimal places.
Sol: Let f  x   3 x  cos x  2  0
f  x  0   3  0   cos  0   2  0  1  2  3  0  ' ' ve 
f  x  1  3 1  cos 1  2  0.459697  0  ' ' ve 
.in
ing
This equation f  x   3 x  cos x  2  0 can be written as
3 x  cos x  2  0
eer
i.e.,3 x  2  cos x
2  cos x
  g  x  , say 
gin
i.e., x 
3
2  cos x
 g  x 
En
3
d
Therefore, g   x    g  x  
dx 
arn
d  2  cos x   1  d 1  sin x

 
dx  3    3  dx  2  cos x    3   0  sin x   3
    
Le
 sin x sin x
 g x  
3 3
w.
sin  0 
g   x  0   0 1
ww
At x = 0, 3
sin 1
g   x  1   0.2804 1
At x = 1, 3
 g   x   1 in (0,1).
Let x0  0  we hvae to take x0  1,∵ f  x  1  f  x  0  
 

5MA8491 Numerical Methods
2  cos  x0  2  cos  0  2  1
.
x1  g  x0     1
3 3 3
2  cos  x1  2  cos 1
x2  g  x1     0.84677
3 3
2  cos  x2  2  cos  0.84677 
x3  g  x2     0.88747
3 3
2  cos  x3  2  cos  0.84747 
x4  g  x3     0.87713
3 3
2  cos  x4  2  cos  0.87713 
x5  g  x4     0.87979   x5  x6 
3 3  rd

  at 3 decimal, 
.in
2  cos  x5  2  cos  0.87979   
x6  g  x5     0.87910   so stop the method 
3 3 
ing
 The root of the given equation 3 x  cos x  2  0 is 0.879
CW.5. Solve the equationcos x  3 x  1 by iteration method upto 5 decimal.

eer
{Solution : x  0.60710 }
HW.6. Solve the equation 2sin x  x by fixed point iteration method.
{Hint : g  x  1  1, so use g  x  1.5   1& g  x  2   1 ,
  
gin
Soln. : x  1.89549 }
x
Eg.7. Solve the equation e  3x  0 by iteration method upto 3 decimal.
En
Let f  x   e  3 x  0
x
Sol:
arn
f  x  0   e0  3  0   1  0  ' ' ve 
f  x  1  e1  3 1  0.2817  0  ' ' ve 
Le

w.
This equation f  x   e  3 x  0 can be written as

x
ww
3x  e x
ex
i.e., x    g  x  , say 
3 
d d  ex  ex
Therefore, g   x    g  x      
dx dx  3  3
ex ex
 g x  
3 3

e0 1
g   x  0    0.333  1
At x = 0, 3 3
e1
g   x  1   0.906  1
At x = 1, 3
 g   x   1 in (0,1).
Let x0  0 , use xn 1  g  xn 
e x0 e0
x1  g  x0     0.33333
.in
3 3 e x6 e0.59983
x7    0.60727
e x1 e0.33333 3 3
x2    0.46520
ing
3 3 e x7 e0.60727
x8    0.61180
e x2 e0.46520 3 3
x3    0.53078
e x8 e0.61180
eer
3 3
x9    0.61458
e x3 e0.53078 3 3
x4    0.56675
e x9 e0.61458
gin
3 3
x10    0.61629
e x4 e0.56675 3 3
x5    0.58751
3 3 e x10 e0.61629
x11    0.61735
En
e x5 e0.58751 3 3
x6    0.59983
3 3
arn
e x11 e0.61735 
x12  g  x11     0.61800   x12  x13 
3 3  rd
Le
x12 0.61800
e e  
x13  g  x12     0.61840   so stop the method 
3 3 
w.
 x
The root of the given equation e  3x  0 is 0.618
Eg.8. Solve the equation 2 x  log x  7 by iteration method upto 5 decimal.
ww
Sol: Let f  x   2 x  log x  7  0

f  x  0   2  0   log10  0   7  Undefined 
f  x  1  2 1  log10 1  7  5  0  ' ' ve 
f  x  2   2  2   log10  2   7  3  0  ' ' ve 
f  x  3  2  3  log10  3  7  1  0  ' ' ve 
f  x  4   2  4   log10  4   7  0.39  0  ' ' ve 

.
This equation f  x   2 x  log x  7  0 can be written as

2 x  7  log x
7  log x
i.e., x    g  x  , say 
2
7  log x
 g  x 
2
1
d d  7  log x  x 1
Therefore, g   x    g  x       2  2x
dx  dx  2 
.in
1 1
 g x  
2x 2x
ing
1 1
g   x  3    0.166  1
At x = 3, 2  3 6
eer
1 1
g   x  4    0.125  1
At x = 4, 2  4 8
gin
 g   x   1 in (3,4).
Let x0  4
En
arn
7  log  x0  7  log  4 
x1  g  x0     3.80103
Le
2 2
7  log  x1  7  log  3.80103
x2  g  x1     3.78995
w.
2 2
7  log  x2  7  log  3.78995 
x3  g  x2     3.78932
ww
2 2
7  log  x3  7  log  3.78932   x4  x5 
x4  g  x3     3.78928   
2 2  th
7  log  x4  7  log  3.78928   
x5  g  x4     3.78928  so stop the method 
2 2 

The root of the given equation 2 x  log x  7 is 3.78928
HW.9. Solve the equation 3 x  log10 x  6 by fixed point iteration method.
{ Solution. : x  2.1080 }

Topic 3 : Newton’s method

(Newton–Raphson method or method of tangents)
Newton–Raphson method formula is

f  xn 
xn 1  xn     xn  , where n  0,1, 2,...
f '  xn 
Part A:
2. State the order of convergence and convergence condition for
Newton–Raphson method.
Sol: Order of convergence is 2. Convergence condition is
2
f  x  f ''  x   f '  x 
.in
3. Find an iterative formula to find N , where N is positive number.
ing
Sol: Let x  N
2 2
 x  N  x N 0
f ' x   2x
eer
Let f  x   x  N 
2
By Newton’s formula
f  xn  x 2  N 2 x 2n  x 2n  N x 2n  N
gin
xn 1  xn   xn  n  
f '  xn  2 xn 2 xn 2 xn
En
4. Derive Newton’s algorithm for finding the pth root of a number N.

1/ p p
Sol: Let x  N  N
arn
p
 x N 0
Let f  x   x  N  f '  x   px
p p 1
Le
By Newton’s formula
xp  N px p  x p  N  p  1 x np  N
w.
n n n
xn 1  xn   
px p 1 px p 1
px p 1
ww
5. Establish an iteration formula to find the reciprocal of a positive

number N by Newton–Raphson method.
1  1
x ∵ The reciprocal of a positive number N is 
Sol: Let N  N 
1 1
N   N 0
 x x
1 1
f  x   N f ' x    2
Let x  x
By Newton’s formula ,

1
N
xn  1 
xn 1  xn   xn    N  xn2  xn  2  Nxn 
1  xn 
 2
xn
3
6. Locate the negative root of x  2 x  5  0 , approximately.
Let f  x   x  2 x  5  0
3
Sol:
f  1  1  2  5  6  ve 
f  2   8  4  5  1 ve 
f  3  27  6  5  16  ve 
 Root lies between –2 and –3.
.in
and root is closer to –2 since f  2   f  3
7. Evaluate 12 applying Newton formula.
Sol: Let x  12
2 2
 x  12  x  12  0 ing
eer
f ' x   2x
Let f  x   x  12 
2
f  3  9  12  3  ve 
gin
f  4   16  12  4  ve 
 Root lies between 3 and 4.
and root is closer to 3 since f  3  f  4 
En
Take x0  3
arn
f  xn 
xn 1  xn 
Newton’s formula is f '  xn 
Le
f  x0  f  3  32  12
x1  x0   3  3  3.5
f '  x0  f '  3 2  3
w.
f  x1  f  3.5   3.52  12
x2  x1   3.5   3.5   3.4642
f '  x1  f '  3.5  2  3.5 
ww

The root is 3.4642
Part B:
Eg.10. Using Newton–Raphson method, find a ‘+’ve root correct to 3
3
decimal places for the equation x  x  2  0 .
Sol: Newton–Raphson method formula is
f  xn 
xn 1  xn  , where n  0,1, 2,...
f '  xn 

Let f  x   x  x  2
3
 f  x   3 x 2  1
f  x  0   03  0  2  2  0  '  've 
f  x  1  1  1  2  2  0
3
 '  've 
f  x  2   23  2  2  4  0  ' ' ve 
 A root lies between 1 and 2.
Take x0  1
f  xn 
xn 1  xn  ,
Now, put n = 0 in f '  xn  we get
.in
f  x0   x0   x0  2
3 13  1  2 
x1  x0   x0  1   2
ing
f '  x0  3  x 2  1 3 12  1
 0   
f  x1   3
  
 x1  1
 x   x1  2  2  
3 2 2 2
  1.63636
eer
x2  x1 
f '  x1  3  x 2  1 3  2 2  1
 1   
f  x2   x2  x2 2
3
1.6363631.636362
gin
x3  x2   x2   1.63636   1.53039
f '  x2  3  x 2 1 3 1.63636 2  1
 2   
f  x3   x3  x32
3
1.5303931.530392
En
x4  x3  x3  1.53039   1.52144

f '  x3  3  x 2 1 3 1.53039 2  1
 3
  
arn
f  x4   x4  x4 2
3
1.5214431.52144 2
x5  x4   x4   1.52144   1.52138
f '  x4  3  x 2 1 3 1.52144 2  1
 4   
Le
f  x5   x5  x5 2
3 3
1.52138  1.52138 2
x6  x5   x5   1.52138   1.52138
f '  x5  3  x 2 1 3 1.52138 2  1
w.
 5
  
 A root of given equation is 1.521
ww
x3  5 x  3  0 by Newton–Raphson method.
CW.11. Solve the equation
{Solution. : x3  x4  0.65662 }
4
x  x  9  0 by Newton–Raphson method.
HW.12. Solve the equation
{Solution. : x6  x7  1.813 upto 3rd decimal}
Eg.13. Using Newton–Raphson method, find a ‘+’ve root correct to 5
decimal places for the equation x tan x  1.28
Sol: Newton–Raphson method formula is

MA8491 Numerical Methods
11 .
f  xn 
xn 1  xn  , where n  0,1, 2,...
f '  xn 
Let f  x   x tan x  1.28

 f  x   x sec 2 x  tan x
f  x  0    0  tan  0   1.28  1.28  0  '  've 
f  x  0.5    0.5  tan  0.5   1.28  1.006  0  '  've 
f  x  1  1 tan 1  1.28  0.2774  0  ' ' ve 
 A root lies between 0.5 and 1.
Take x0  0.5
.in
f  xn 
xn 1  xn  ,
ing
f  x0  x0 tan x0 1.28  0.5 tan  0.51.28
x1  x0   x0   0.5   1.34218
f '  x0  2
x0 sec x0 tan x0  0.5 sec2  0.5tan  0.5
eer
f  x1  1.34218 tan 1.342181.28
x2  x1   1.34218   1.19469
f '  x1  1.34218 sec2 1.34218tan 1.34218
f  x2  1.19469  tan 1.19469 1.28
gin
x3  x2   1.19469   1.04143
f '  x2  1.19469  sec2 1.19469 tan 1.19469 
f  x3  1.04143 tan 1.041431.28
x4  x3   1.04143   0.95512
f '  x3 
En
1.04143 sec2 1.04143tan 1.04143

f  x4   0.95512  tan  0.95512 1.28
x5  x4   0.95512   0.93871
f '  x4    sec2  0.95512 tan  0.95512 
arn
0.95512
f  x5   0.93871 tan  0.938711.28
x6  x5   0.93871   0.93826
f '  x5   0.93871 sec2  0.93871tan  0.93871
f  x6 
Le
 0.93826  tan  0.93826 1.28

x7  x6   0.93826   0.93826
f '  x6   0.93826  sec2  0.93826 tan  0.93826 
w.

x sin x  cos x  0 by Newton–Raphson method.
CW.14. Solve the equation
ww
{Solution. : x4  x5  2.79839 }
2 x
xe
HW.15. Solve the equation  0.5sin x by Newton–Raphson method.
{Solution. : x2  x3  3.12962 upto 5th decimal}
Eg.16. Using Newton–Raphson method, find a root correct to 5 decimal
x
places for the equation e  4 x  0
f  xn 
xn 1  xn 
Sol: Newton–Raphson method formula is f '  xn 
Let f  x   e  4 x  f  x   e  4
x x

f  x  0   e0  4  0   1  0  ' ' ve 
f  x  1  e1  4 1  1.28272  0  '  've 
 A root lies between 0 and 1.
Here f 1  1.28172  f  0   1
 Root is nearer to 0.
Take x0  0
f  xn 
xn 1  xn  ,
f  x0  e x0  4 x0 e0  4  0 
x1  x0   x0  x  0  0.33333
.in
f '  x0  e 0 4 e0  4
f  x1  e x1 4 x e0.33333 4  0.33333
x2  x1   x1  x 1  0.33333   0.35725
ing
f '  x1  e 1 4 e0.33333 4
f  x2  e x2 4 x2 e0.35725 4  0.35725 
x3  x2   x2  x  0.35725   0.35740
f '  x2  e0.35725 4
eer
e 2 4
f  x3  e x3 4 x e0.35740 4  0.35740 
x4  x3   x3  x 3  0.35740   0.35740
f '  x3  e 3 4 e0.35740 4
gin

En
x
CW.17. Solve the equation cos x  xe , by x0  0.5 by Newton–Raphson
method. {Solution. : x2  x3  0.51776 }
arn
2 x  log10 x  7 by Newton–Raphson method.

HW.18. Solve the equation
{Solution. : x5  x6  3.78928 upto 5th decimal}
Le
1
Eg.19. Find the Newton–Raphson formula to find the value of N where
1
w.
N is a real number, hence evaluate 26 correct to 4 decimal places.

ww
1
x
Sol: Let N
1 1
N   N 0
 x x
1 1
f  x   N f ' x    2
Let x  x
By Newton’s formula ,

13 .
1
N
xn  1 
xn 1  xn   xn    N  xn2  xn  2  Nxn 
1  xn 
 2
xn
1
This is the Newton’s–Raphson formula to find the value of N .
1
To find :
26
Put N = 26 and n = 0 in xn 1  xn  2  Nxn  , we get
x1  x0  2  26 x0  (1)
.in
1  1 1 
Take  0.04 as x0 ∵ is nearer to 
25  25 26 
ing
 1  x1  x0  2  26 x0   0.04  2  26  0.04    0.0384
x2  x1  2  26 x1   0.0384  2  26  0.0384    0.03846
eer
x3  x2  2  26 x2   0.03846  2  26  0.03846    0.03846
 A root of given equation is 0.03846 or 0.0385
gin
CW.20. Find the Newton–Raphson formula to find the value of N where

N is a real number, hence evaluate 142 correct to 5 decimal
{Solution. : x2  x3  11.91638 }
En
places.
1
p
HW.21. Find the Newton–Raphson formula to find the value of N where
arn
N is a real number, hence evaluate 3 17 correct to 5 decimal

places. {Solution. : x4  x5  2.57128 }
Le
Anna University Questions:

AU1. Find Newton’s iterative formula for the reciprocal of a number N
w.
1
and hence find the value of23 , correct to five decimal places.
ww
(Anna Univ. Nov./Dec, 2009(ECE)) Ans: x = 0.04347

AU2. Prove the quadratic convergence of Newton’s-Raphson method.
3
Find a positive root of f ( x)  x  5 x  3  0 , using this method.
(May/June, 2007) Ans: x = 0.6566
AU3. Use Newton’s method to find the real root of 3x − cos x − 1 = 0.
(Nov./Dec., 2007) Ans: 0.607102
3
AU4. Obtain the positive root of 2 x  3 x  6  0 that lies between 1 and
2 by using Newton’s-Raphson method.
(May/June, 2007(EEE)) Ans: 1.7838

Topic 4 : Solution of linear system by Gaussian elimination method
Gauss elimination method:
Step1: Write the augmented matrix for the given system of
simultaneous equations
 a11 a12 a13 b1 
 a21 a22 a23 b2 
a 
 31 a32 b33 b3 
Step2: Using elementary row operations reduce the given
matrix into an upper–triangular matrix say
 c11 c12 c13 d1 
 0 c22 c23 d 2 
 0 0 c33 d3 

.in
Step3: By back substitution we get the values for unknowns.
Topic 5 : Solution of linear system by Gaussian Jordan method
Gauss Jordan method:
ing
Step1: Write the augmented matrix for the given system of
simultaneous equations
 a11 a12 a13 b1 
eer
 a21 a22 a23 b2 
a 
 31 a32 b33 b3 
gin
Step2: Using elementary row operations reduce the given

matrix into a diagonal matrix say
 c11 0 0 d1    c11 0 0   diagonal matrix  
 0 c22 0 d 2    
En
  
 0   Here  0 c22 0  is a  or  
 0 c d 3
33
     
  0 0 c33   unit matrix  
arn
Step3: By direct substitution we get the values for unknowns.

(Note: Gauss elimination and Gauss Jordan methods are direct methods.)
Part A:
Le
8. For solving a linear system AX = B, compare Gauss elimination

method and Gauss Jordan method.
w.
Sol:
9. Gauss elimination method Gauss Jordan method
Coefficient matrix is transmitted Coefficient matrix is transmitted
ww
into upper triangular matrix into diagonal matrix or unit matrix

Direct method Direct method
Obtain the solution by back No need of back substitution
substitution method method
State the two difference between direct and iterative
methods for solving system of equations.
Sol: Direct Method Iterative Method
It gives exact value It gives only approximate value
Simple, takes less time Time consuming and labourious
This method determine all the This method determine only one
roots at the same time root at a time

15 .
Part B:
Eg.22.Solve the system of equations by (i) Gauss elimination method (ii)
Gauss Jordan method
3 x1  2 x2  3 x3  6
x1  x2  x3  1
2 x1  5 x2  4 x3  5
Sol: The given system of equations can be written as
x1  x2  x3  1
3 x1  2 x2  3 x3  6
2 x1  5 x2  4 x3  5
Given system of the form AX = B
(i) Gauss elimination method:
.in
The augmented matrix form is
 1 1 1 1 
ing
 A | B    3 2 3 6
 2 5 4 5 
eer
 1 1 1 1 
R  R2  3R1
∼  3  3 2  3 3  3 6  3 2
R  R3  2 R1
gin
 2  2 5  2 4  2 5  2  3
1 1 1 1 
∼ 0 1 0 3
En
0 3 2 3 
arn
1 1 1 1 

∼ 0 1 0 3  R3  R3  3R2
0 3  3 2  0 3  9 
Le
1 1 1 1 
∼ 0 1 0 3
w.
0 0 2 12 
ww
By back substitution to find the solution of the system of equations

2 x3  12  x3  6
 x2  3  x2  3
x1  x2  x3  1  x1  2
  2, 3, 6  is the solution for the system of equations.
(ii) Gauss–Jordan method:
The augmented matrix form is

 1 1 1 1 
   3 2 3 6
A | B 
 2 5 4 5 
 1 1 1 1 
R  R2  3R1
∼  3  3 2  3 3  3 6  3 2
R  R3  2 R1
 2  2 5  2 4  2 5  2  3
1 1 1 1
∼ 0 1 0 3
0 3 2 3 
1 1 1 1
.in
∼ 0 1 0 3 R2  R2   1
ing
0 3 2 3
1  0 1  1 1 0 1 3 
R  R1  R2
∼  0 3  1
eer
1 0
R  R3  3R2
 0 3  3(1) 2  3(0) 3  3(3)  3
1 0 1 4 
gin
∼ 0 1 0 3 
0 0 2 12 
En
1 0 1 4 
R
∼ 0 1 0 3  R3  3
arn
2
0 0 1 6 
1 0 0 2 
Le
∼ 0 1 0 3  R1  R1  R3
0 0 1 6 
w.
 x1  2, x2  3, x3  6   x1 , x2 , x3    2,3,6 
ww
CW23. Solve the system of equation by (i) Gauss elimination method (ii)
Gauss Jordan method
28 x  4 y  z  32  x, y, z    0.99359,1.50697,1.84855
x  3 y  10 z  24
2 x  17 y  4 z  35 {Sol: }
HW24. Solve the system of equation by (i) Gauss elimination method (ii)
Gauss Jordan method
pqrs2  p, q, r , s   1, 0, 1, 2 
p  q  3r  2 s  6
2 p  3q  r  2 s  7
p  2q  r  s  2 { Sol: }

17 .
AU5. Solve the given system of equations by Gauss elimination method:

 x1  x2  10 x3  35.61,10 x1  x2  x3  11.19, x1  10 x2  x3  20.08
(Nov./Dec. 2008(ECE))
AU6. Solve by Gauss Jordan method, the following system
10x + y − z = 11.19; x + 10y + z = 20.08; − x + y + 10z = 35.61.
(Nov./Dec., 2009(ECE))
AU7. Apply Gauss Jordan method to solve the equations
x + y + z = 9; 2x − 3y + 4z = 13; 3x + 4y + 5z = 40.
11 19 59
x  ,y  ,z 
(Nov./Dec., 2007) Ans: 12 6 12
AU8. Using Gauss-Jordon method, solve the following system of
equations 2x − y + 3z = 8; − x + 2y + z = 4; 3x + y − 4z = 0
.in
(Nov./Dec., 2007(EEE)) Ans: x = 2, y = 2, z = 2
ing
Topic 6 : Gauss-Seidel method(Iterative method):
Step1: Let the system of equations be
a11 x  a12 y  a13 z  b1
eer
a21 x  a22 y  a23 z  b2
a31 x  a32 y  a33 z  b3 with diagonally dominant
Step2: The above system should write into the form
gin
1
x  b1  a12 y  a13 z  ...(1)
a11
1
 b2  a21 x  a23 z  ...(2)
En
y
a22
1
z  b3  a31 x  a32 y  ...(3)
arn
a33
 0  0
Step3: Start with the initial values y  0, z  0 for y, z and get
Le
x  from the first equation.

1
x  
1

1

b1  a12 y    a13 z  
0 0
w.
 (1) becomes a11

1  0 1
ww
Step4: Using this x in (2), we use z for z and x for x

 0
instead of x , we get
 (2) becomes
y  
1 1

a22
1

b2  a21 x   a23 z  
0
1 1
Step5: Substitute x , y for x, y in the third equation.
 (3) becomes
z  
1 1

a33
1

b3  a31 x   a32 y  
1

Step5: To find the values of unknowns, use the latest available
r r r
values on the right side. If x , y , z are the rth iterate
values, then the next iteration will be
x  
r 1
1
a11

b1  a12 y    a13 z  
r r
y  r 1
 
1
a22

b2  a21 x   a23 z  
r 1 r
z  r 1
 1
a33
b3  a31 x   a32 y  
r 1

r 1
Step6: This process of continued until the convergence is

confirmed.
.in
Part A:
10. Write the first iteration values of x, y, z when the equations
27 x  y  z  8, x  1 y  2 z  72, x  y  z  110 are solved by
ing
Gauss–Seidel method.
Sol:
eer
1 85
x   85  6 y  z  
27 27 = 3.14815 [putting y = z = 0]
1 1
y   72  6 x  2 z   72  6  3.14815   2  0    3.54074
gin
15 15
[putting x= 3.14815 , z = 0]
En
1 1
z  110  x  y   110  3.14815  3.54074  20.66222
5 5
[putting x = 3.14815 , y = 3.54074 ]
arn
Part B:
Eg.25. Solve by Gauss-Seidel iterative method
Le
x  y  54 z  110
27 x  6 y  z  85
6 x  15 y  2 z  72
w.
Sol: This can be written in diagonally dominant matrix as

27 x  6 y  z  85  (1)
ww
6 x  15 y  2 z  72  (2)
x  y  54 z  110  (3)
1
(1)  x  85  6 y  z   (4)
27
1
(2)  y   72  6 x  2 z   (5)
15
1
(3)  z  110  x  y   (6)
54
Iteration 1:

19 .
Put y  z  0 in (4)
1
(4)  x  85  6  0   0   3.14815
27 
 1
(5)  y  72  6  3.14815   2  0    3.54074
15
1
(6)  z  110  3.14815  3.54074   1.91317
54
Iteration 2: [Substitute the latest values of x, y, z in (4), (5), (6)]
1
x 85  6  3.54074   1.91317   2.43218
27 
1
y  72  6  2.43218   2 1.91317    3.57204
.in
15
1
z  110  2.43218  3.57204  1.92585
54
ing
1
Iteration 3: x 85  6  3.57204   1.92585  2.42569
27 
1
eer
y  72  6  2.42569   2 1.92585    3.57294
15
1
z  110  2.42569  3.57294  1.92595
gin
54
1
Iteration 4: x 85  6  3.57294   1.92595  2.42549
27 
En
1
y  72  6  2.42549   2 1.92595    3.57301
15
1
z  110  2.42549  3.57301  1.92595
arn
54
1
Iteration 5: x 85  6  3.57301  1.92595  2.42548
27 
Le
1
y  72  6  2.42548   2 1.92595    3.57301
15
w.
1
z  110  2.42548  3.57301  1.92595
54
ww

Solution is  x, y, z    2.42548, 3.57301, 1.92595 
Note : Iterations of the above given problem are given as follows:
Iteration x y z
1 3.14815 3.54074 1.91317
2 2.43218 3.57204 1.92585
3 2.42569 3.57294 1.92595
4 2.42549 3.57301 1.92595
5 2.42548 3.57301 1.92595
CW.26. Solve by Gauss-Seidel iterative method

8 x  3 y  2 z  20
6 x  3 y  12 z  35
4 x  11 y  z  33
Sol:  x, y, z    3.016,1.985, 0.911
HW.27. Apply Gauss–Seidel method to solve the system of equations
20 x  y  2 z  17,3x  20 y  z  18, 2 x  3 y  20 z  25
Sol:  x, y, z    0.999969  1, 1.0000064  1,1.000002  1
AU9. Solve the following system by Gauss- Seidal method :
28x + 4y − z = 32; x + 3y + 10z = 24; 2x + 17y + 4z = 35
(May./June, 2007)Ans: x=0.9936,y=1.507,z=1.8486
AU10.By using Gauss-Seidel method, solve the system of equations
6x + 3y + 12z = 35; 8x − 3y + 2z = 20; 4x + 11y − z = 33.
.in
(May./June, 2007(EEE))
AU11.By using Gauss-Seidel iteration method, solve the following
ing
system of equations upto four decimals.
10x − 2y − z − w = 3; −2x + 10y − z − w = 15;
−x − y + 10z − 2w = 27; − x − y − 2z + 10w = −9.
eer
(Nov./Dec, 2009(IT))Ans: x = 3.017, y = 1.986, z = 0.91
Topic 7 : Inverse of a matrix by Gauss Jordan Method:

gin
Given Matrix An  n ,
Row
 An n | I n n  ~  I n  n | A1n  n 
 
En
Form Operations
Part A:
 
arn
1 3
11. Find the inverse of a matrix 2 7 by Gauss–Jordan method.
1 3:1 0 
Le
Sol: Consider [A | I] =  2 7:0 1 

∼ 1 3: 1 0  R2  R2  2 R1
w.
0 1: 2 1 
∼ 1 0: 7 3 R1  R1  3R2
0 1: 2 1 
ww
 A1   7 3
 2 1 
Part B:
Eg.28.Using Gauss–Jordan method, find the inverse of the matrix
1 0 1
3 4 5 
 
0 6 7 

21 .
1 0 1
A   3 4 5 
Sol: Let 0 6 7 
1 0 1 1 0 0
Consider  A | I    3 4 5 0 1 0  Apply R2  R2  3R1
0 6 7 0 0 1 
 1 0 1 1 0 0 

∼ 3  3 1 4  3  0  5  3  1 0  3 1 1  3  0  0  3  0  
 0 6 7 0 0 1 
1 0 1 1 0 0 
.in
R
∼ 0 4 8 3 1 0  R2  2
4
0 6 7 0 0 1 
1 0 1 1 0 0
∼ 0 1 2 3 / 4 1 / 4 0  R3  R3  6  R2  ing
eer
0 6 7 0 0 1 
1 0 1 1 0 0 
gin
 
∼ 0 1 2 3 / 4 1/ 4 0 
0 6  6 1 7  6  2  0  6  3 / 4  0  6 1 / 4  1  6  0  
En
1 0 1 1 0 0
R
∼ 0 1 2 3 / 4 1 / 4 0  R3  3

5
arn
0 0 5 9 / 2 3 / 2 1 
1 0 1 1 0 0 
R1  R1  R3
∼ 0 1 2 3 / 4 1 / 4 0 
Le
R  R2  2  R3 
0 0 1 9 / 10 3 / 10 1 / 5 2
w.
 1 0 00 1  1 1 9 / 10  0 3 / 10  01 / 5  

 
∼ 02  0  12  0  22 1  3 / 4 2  9 / 10  1 / 4 2  3 / 10  02 1 / 5  
ww
 0 0 1 9 / 10 3 / 10 1 / 5 
1 0 0 1 / 10 3 / 10 1 / 5 
∼ 0 1 0 21 / 20 7 / 20 2 / 5   I | A1 
 
0 0 1 9 / 10 3 / 10 1 / 5 
 1 3 1
1 
 A1   21/ 2 7 / 2 4 
10
 9 3 2 

1 1 3
 1 3 3
CW.29. Find the inverse of a matrix  2 4 4  by Gauss–Jordan
1 12 4 6 
A1   5 1 3
4  1 1 1
method. Soln. :
HW.30. Using Gauss–Jordan method, find the inverse of the matrix
2 2 6  12 4 6 
 2 6 6  1 
 
1
A  1 5 3
56 
 4 8 8 Soln. :  5 3 1
1 2 1
.in
4 1 0 
AU12.Find the inverse of the matrix  2 1 3  by using Gauss–Jordan
ing
 1/ 5 1/ 3 1/ 15
A1   4 / 5 1/ 3 4 / 15 
method. (M/J,'07(EEE)) Ans:  2 / 5 1/ 3 7 / 15 
eer
AU13.Find the inverse of the matrix by Gauss-Jordan method:
4 1 2   4 / 3 2 7/3 
gin
A   2 3 1 A1   5 / 3 2 8 / 3 
1 2 2  (M/,'07) Ans:  7 / 3 3 10 / 3
En
Topic 8 : Eigen value of a matrix by power method(Iterative method)

Suppose a given square matrix is A.
arn
1 
X1   0 
Let 0  be an arbitrary initial eigen vector of the matrix A.
Le
 value1   numerically largest value   1 

AX1  value2    of matrix A say  value2 / value1  1  X 2 
 value3    
value1   value3 / value1
w.
 value1   numerically largest value   value1/ value2 

AX 2  value2    of matrix A say  1   2  X 3 
ww
  value3 / value2 
 value3   value2  
⋮
AX n  n  X n 1   If n =n 1 and X n 1  X n  2 are same
AX n 1  n 1  X n  2  upto required decimals, then stop iteration.
Dominant eigen value = n  or  n 1 and the corresponding

eigen vector = X n  or  X n 1

23 .
To find the smallest eigen value of A

 1 
Y1  0 
Form B = A – I & 0  be an arb. eigen vector of the matrix B.
 value1   numerically largest value   1 
BY1  value2    of matrix B say  value2 / value1  1 Y2 
   value3 / value1
 value3   value1  
 value1   numerically largest value   value1/ value2 
BY2  value2    of matrix B say  1   2 Y3 
 value3   value2  value3 / value2 
 
⋮
.in
BYn  n Yn 1   If n =n 1 and Yn 1  Yn  2 are same
BYn 1  n 1 Yn  2  upto required decimals, then stop iteration.
ing
Smallest eigen value of B
 
 The smallest eigen value of A =   Dominant eigen value of A
eer
To find third eigen value of A:
1  2  3  Sum of the main diagonal elements
gin
Part A:
12. What type of eigen value can by obtained by using power method.
Sol: We can obtain dominant eigen value of the given matrix
En
13. Write down all possible initial vectors of a matrix 2 x 2 order.

T T T
Sol: The initial vectors are (0,1) ,(1,0) ,(1,1) .
arn
A   4 1
14. Find the dominant eigen value of 1 3 by power method
1
Le
upto two decimals and choose 1 as the initial eigen vector.
X1  1
w.
Sol: Let 1

AX1   4 1 1  5   5  1 
ww
 5X 2
1 3 1  4  0.8
AX 2   4 1  1    4.8  4.8  1   4.8 X 3
1 3 0.8 3.4  0.71
AX 3   4 1   1    4.71   4.71  1   4.71X 4
1 3 0.71 3.13 0.67 
AX 4   4 1   1    4.67   4.67  1   4.67 X 5
1 3 0.67  3.01  0.65

AX 5   4 1  1    4.65  4.65  1   4.65 X 6
1 3 0.65  2.95 0.63
AX 6   4 1   1    4.63   4.63  1   4.63 X 7
1 3 0.63  2.89  0.62 
AX 8   4 1   1    4.62   4.62  1   4.62 X 8
1 3 0.62   2.86  0.62 

 The eigen value = = 4.62 and
 1 
the corresponding eigen vector = X = 0.62 
Part B:
Eg.31.Find the dominant eigen value and eigen vector of the matrix
 4 5
.in
 1 2  by power method.
1 
X1   
ing
Sol: Let 0  be an arbitrary initial eigen vector.
 4 5 1   4  1   5  0    4   1 
eer
AX1            4    4 X 2 , say
 1 2  0   1  1   2  0    1   0.25
4 5  1    41   5 0.25   2.75   2.75  1   2.75 X
gin
AX 2 
 1 2  0.25  1  1   2   0.25  0.5  0.18182 3
 
 4 5  1   3.0909   1 
En
AX 3          3.0909    3.0909 X 4
1 2   0.18182   0.63636    0.20588 
 4 5  1   2.9706   1 
arn
AX 4          2.9706    2.9706 X 5
1 2   0.20588  0.58824   0.06666 
 4 5  1   3.6667   1 
Le
AX 5     0.06666    0.86668    3.6667  0.23637   3.6667 X 6

 1 2      
 4 5  1   2.81815  1 
w.
AX 6         2.81815    2.81815 X 7
1 2   0.23637   0.52726   0.18709 
 4 5  1   3.06455  1 
ww
AX 7         3.06455    3.06455 X 8
1 2   0.18709   0.62582   0.19579 
 4 5  1   3.02105  1 
AX 8         3.022105    3.022105 X 9
 1 2   0.19579   0.60842   0.20133
Dominant Eigen value =  = –3 and the corresponding
 1  1 
Eigen vector = X =  0.19529   0.2 

25 .
CW.32. Find by power method, the largest eigen value and the
1 3 1
3 2 4
corresponding eigen vector of the matrix  1 4 10  .
 0.02404 
0.42582 
Sol: Dom. Eigen value = = 11.72 and Eigen vector = X =  1 
1 6 1 
1 2 0 
Eg.33. Find all the eigen values of the matrix 0 0 3  by power method.
Sol: Dominant Eigen value =  = 4
Smallest Eigen value = X = –1
.in
Other Eigen value = 3
AU14. Obtain by power method the numerically largest eigen value of the
 15 4 3   
ing
 10 12 6 
matrix  20 4 21 (N/D., 2007) Ans: = −19.977 −20
eer
1 2 3 
0 4 2 
AU15. Find the all eigen values of the matrix by power method 0 0 7 
gin
(Nov./Dec., 2007(EEE)) Ans:  = −4, 1, 7

AU16. Find the dominant Eigen value and the corresponding Eigen vector
1 6 1  
En
1 2 0 
of the matrix 0 0 3 (May./June, 07)Ans: = −1, 3, 4
arn
AU17. Find, by power method, the largest eigen value and the eigen vector
 25 1 2 
 1 3 0
of the matrix  2 0 4  (May./June, 2007(EEE))
Le
Ans:  = 25.1821 ; X = (1, 0.0451, 0.0685)T

w.
AU18. Solve by power method, to find the dominant Eigen value for the
1 1 3 
1 5 1
ww
following matrix 3 1 1 (Nov./Dec., 2009(ECE)) Ans: =

AU19. Determine, by power method, the largest eigenvalue of the matrix
 2 1 0  
 1 2 1
 0 1 2  (Nov./Dec., 2009(IT)) Ans: =
Topic9: Eigenvalue of a matrix by Jacobi method for symmetric matrix

For a given symmetrical matrix A, find a dominant element aij in
off–diagonal element(i.e., element not in diagonal),
Change element elements as

aij =  sin  , a ji = sin  , aii  a jj  cos 
Change the remaining diagonal elements as 1
Change the remaining off–diagonal elements as 0
 The first rotation matrix is
i th col. j th col.
1 0 0 ⋯ 0 ⋯ 0 ⋯ 0
0 1 0 ⋯ 0 ⋯ 0 ⋯ 0
⋮ 
th 0 0 0 ⋯ cos  ⋯  sin 
S1  i row ⋯ 0
⋮ ⋮ ⋮ 
th
j row  0 ⋯ ⋯ ⋯ sin  ⋯ cos  ⋯ 0
⋮ 
0 ⋯ ⋯ ⋯ 0 ⋯ 0 ⋯ 1 

.in
1  2aij 
 tan 1   , if aii  a jj
ing
2  aii  a jj 


    , if a  0 
 4 ij 
eer
   , if aii  a jj
 4 , if aij  0 
Use 
gin
 
  
Here 4 4
T
and form transformation as B1  S1 A S1
En
If B1  
a diagonal matrix, then S1 is its corresponding vector.
not a diagonal matrix, then form seond rotation matrix S2 .
arn
T
and form transformation as B2  S 2 A S2
a diagonal matrix, then S2 is its corresponding vector.

Le
If B2 
not a diagonal matrix, then form seond rotation matrix S3 .
…
w.
continue till Bn as diagonal matrix with

eigen values as its diagonal elements.
ww
Now eigen vectors obtained as follows:

Form S  S1S2 S3 ...Sn with columns as eigen vectors of
eigen values.
Part A:
15. Using Jacobi’s method find the eigne values and eigen vectors of
1 2 
 2 1  .
1 2 
Sol: Let A =  2 1 
The largest off–diagonal element a12  a21  2

27 .
S1  cos   sin  
The rotation matrix  sin  cos  
a11  a22 

Since , we have 4
  
cos 4  sin 4  1/ 2 1/ 2 
S1   
   1/ 2 1/ 2 
 sin cos 
  4 4 
Then the transformation gives
   
B1  S1 A S1   1/ 2 1/ 2  1 2  1/ 2 1/ 2    3 0 
T
 1/ 2 1/ 2  
 2 1 
 1 / 2 1/ 2  0 1
.in
This is a diagonal matrix. So, the eigen values are 3, –1 and the
corresponding eigen vectors are the columns of S1 i.e.,
ing
1/ 2   1/ 2 
1/ 2  ,  1/ 2 
   .
eer
1 5 2
 5 4 7 
gin
16. Using Jacobi’s method find the rotation matrix of  2 7 4  .

1 5 2
 5 4 7 
En
Sol: Let A =  2 7 4 
The numerically largest off–diagonal element a23  a32  7
arn
1 0 0 
S1  0 cos   sin  
The rotation matrix 0 sin  cos  
Le
w.
Part B:
Eg.34.Find all the eigen values and eigen vectors of the matrix
ww
 1 2 2 
A 2 3 2
 
 2 2 1 
by Jacobi method.
Sol: Given matrix is symmetric matrix.
Numerically largest element in off diagonal element is 2 in a13 &
in a31 so multiply with  sin  & sin  and other elements
a11 & a33 are both multiplied by cos  remaining diagonal a22
with ‘1’ and a12 , a21 , a23 , a32 with ‘0’.

cos  0  sin  
 Suitable orthogonal matrix is S1   0 1 0 
 sin  0 cos  
1  2a13  1 1  2  2   1   
 tan 1    tan    
2  a11  a33  2  11  2  2  4
cos  / 4  0 sin  / 4   1 / 2 0 1 / 2 
 S1   0 1 0  0 1 0 
 sin  / 4  0 cos  / 4   1 / 2 0 1 / 2 
   
 1/ 2 0 1/ 2   1 2 2  1/ 2 0 1/ 2 
Now B1  S1T AS1   0 1 0  2 3 2  0 1 0 
 1/ 2 0 1/ 2   2 2 1

1/ 2 0 1/ 2 
    
.in
3 2 0 
 2 3 0 
ing
 0 0 1
cos   sin  0 
Here S2   sin  cos  0 
 0 1 
eer
0
1  2a12  1 1  2  2   1 1  
Now   tan 1    tan       
1
  tan 
2 a  a
 11 22  2  3  3  2 2 2  4
gin
1/ 2 1/ 2 0 
 S2  1/ 2 1/ 2 0 
 0 1 
En
0
 
 1/ 2 1/ 2 0   3 2 0  1/ 2 1/ 2 0 
arn
B2  S2T B1S2   1/ 2 1/ 2 0   2 3 0  1/ 2 1/ 2 0 

 0 0 1   0 0 1  0 0 1 
   
5 0 0
Le
 0 1 0  , a diagonal matrix with 5,1, 1 as eigen values

0 0 1
w.
To find Eigen vectors:

ww
X1 X2 X3
1/ 2 0 1/ 2  1/ 2 1/ 2 0   1/ 2 1/ 2 1/ 2 
S1S2   0 1 0  1/ 2 1/ 2 0   1/ 2 1/ 2 0 
1/ 2 0 1/ 2   0 0 1   1/ 2 1/ 2 1/ 2 
 
     
1/ 2   1/ 2   1/ 2 
 Eigen vectors are X1  1 / 2  , X 2  1/ 2  , X 3  0 
1/ 2   1/ 2  1/ 2 
     
for corresponding eigen values 5,1, 1.

29 .
CW.35. Find all the eigen values and eigen vectors of the matrix
1/ 2 1 2 
A 1 3/ 2 1 
 
 2 1 1/ 2 
by Jacobi method.
Sol: Eigen values = 5 / 2, 1/ 2,  1/ 2 and
 1 / 2   1 / 2   1 / 2 
1 / 2  , 1 / 2  ,  0 
 1 / 2   1 / 2   1 / 2 
Eigen vectors =      
HW.36. Find all the eigen values and eigen vectors of the matrix
 2 1 1 
A   1 2 1
.in
 1 1 2  by Jacobi method.
Sol: Eigen
Corresponding eigen vectors
ing
Values
1,4,1 1/ 6  1   1/ 3   1  1   1/ 2   1
 2 / 6  or  2  ,  1/ 3  or  1  ∼  1 ,  0  or  0 
eer
             
1/ 6  1   1/ 3   1  1   1/ 2   1 
gin
En
arn
Le
w.
ww

Unit I : Assignments Problems
1. Solve the equation 3 x  cos x  2  0 by iteration method.

Sol: x = 0.879
2. Solve the equation 2 x  log x  7 by iteration method.
Sol: x = 3.789
3. Find Newton’s iterative formula for the reciprocal of a number N
1
and hence find the value of23 , correct to five decimal places.
Ans: x = 0.04347
4. Use Newton’s method to find the real root of 3x − cos x − 1 = 0.
.in
Ans: 0.607102
5. Solve the system of equation by (i) Gauss elimination method (ii)
Gauss Jordan method
ing
3 x1  2 x2  3 x3  6
x1  x2  x3  1
2 x1  5 x2  4 x3  5
eer
Sol:  x1 , x2 , x3    2,3,6 
6. Solve the given system of equations by (i) Gauss elimination
gin
method (ii) Gauss Jordan method

 x1  x2  10 x3  35.61,10 x1  x2  x3  11.19, x1  10 x2  x3  20.08
7. Solve the following system by Gauss- Seidal method :
En
28x + 4y − z = 32; x + 3y + 10z = 24; 2x + 17y + 4z = 35

Ans: x=0.9936,y=1.507,z=1.8486
arn
8. By using Gauss-Seidel method, solve the system of equations

6x + 3y + 12z = 35; 8x − 3y + 2z = 20; 4x + 11y − z = 33.
1 2 1
Le
4 1 0 
9. Find the inverse of the matrix  2 1 3  by using Gauss–Jordan
w.
 1/ 5 1/ 3 1/ 15

A1   4 / 5 1/ 3 4 / 15 
method. Ans:  2 / 5 1/ 3 7 / 15 
ww
10. Find the inverse of the matrix by Gauss-Jordan method:

4 1 2   4 / 3 2 7/3 
A   2 3 1 A1   5 / 3 2 8 / 3 
1 2 2  Ans:  7 / 3 3 10 / 3
11. Obtain by power method the numerically largest eigen value of the
 ≃
 15 4 3 
 10 12  6 
matrix  20 4 21 Ans: = −19.977 −20

31 .
12. Find the all eigen values of the matrix by power method
1 2 3  
0 4 2 
0 0 7  Ans: = −4, 1, 7
13. Find the dominant Eigen value and the corresponding Eigen vector
1 6 1  
1 2 0 
of the matrix 0 0 3 Ans: = −1, 3, 4
14. Find all the eigen values and eigen vectors of the matrix
 1 2 2 
A 2 3 2
 
.in
 2 2 1 
by Jacobi method.
Sol: Eigen values = 5, 1, –1 and
ing
 1 / 2   1 / 2   1 / 2 
1 / 2  , 1 / 2  ,  0 
 1 / 2   1 / 2   1 / 2 
eer
1/ 2 1 2 
gin
A 1 3/ 2 1 
 
 2 1 1/ 2 
by Jacobi method.
En
Sol: Eigen values = 5 / 2, 1/ 2,  1/ 2 and

 1 / 2   1 / 2   1 / 2 
1 / 2  , 1 / 2  ,  0 
arn
 1 / 2   1 / 2   1 / 2 
Le
 2 1 1 
A   1 2 1
 1 1 2  by Jacobi method.
w.
Sol: Eigen values = 1, 4, 1 and

ww
1/ 6   1/ 3   1/ 2 

 2 / 6  ,  1/ 3  ,  0 
   
1/ 6   1/ 3   1/ 2 
Eigen vectors = 

UNIT II: INTERPOLATION AND APPROXIMATION

Lagrangian Polynomials – Divided differences – Interpolating with a cubic
spline – Newton’s forward and backward difference formulas.
Topic 1 : Lagrangian Polynomials(Equal and unequal intervals)

Lagrange’s interpolation formula(x given, finding y in terms of x)
Let y =f(x) be a function which takes the values y = y0, y1,…, yn
corresponding to x = x0, x1,…, xn.
( x  x1 )( x  x2 )… ( x  xn )
y  y ( x)  f ( x)  y0
( x0  x1 )( x0  x2 )… ( x0  xn )
( x  x0 )( x  x2 )… ( x  xn )
.in
 y1  …
( x1  x0 )( x1  x2 )… ( x1  xn )
( x  x0 )( x  x1 )… ( x  xn 1 )
ing
 yn
( x2  x0 )( x2  x1 )… ( x  xn 1 )
Inverse Lagrange’s interpolation formula(y given, finding x in
eer
terms of y)
( y  y1 )( y  y2 )… ( y  yn )
x  x( y )  f ( y )  x0
( y0  y1 )( y0  y2 )… ( y0  yn )
gin
( y  y0 )( y  y2 )… ( y  yn )
 x1  …
( y1  y0 )( y1  y2 )… ( y1  yn )
En
( y  y0 )( y  y1 )… ( y  yn 1 )
 xn
( y2  y0 )( y2  y1 )… ( y  yn 1 )
arn
Note:
Lagrange’s interpolation formula can be used for
equal and unequal intervals.
Le
Part A:
1. What is the assumptions we make when Lagrange’s formula is
used?
w.
Sol: Lagrange’s interpolation formula can be used whether the

values of x, the independent variable are equally spaced or not
ww
whether the difference of y become smaller or not.

2. What is the disadvantage in practice in applying Lagrange’s
interpolation formula?
Sol: Though Lagrange’s formula is simple and easy to
remember, its application is not speedy. It requires close attention
to sign and there is always a chance of committing some error due
to a number of positive and negative signs in the numerator and the
denominator.

2 UNIT II: Interpolation And Approximation
3. What is ‘inverse interpolation’?

Sol: Suppose we are given a table of values of x and y. Direct
interpolation is the process if finding the values of y corresponding
to a value of x, not present in the table. Inverse interpolation is the
process of finding the values of x corresponding to a value of y, not
present in the table.
4. Construct a linear interpolating polynomial given the points
( x0 , y0 ) and ( x1, y1 ).
( x  x1 ) ( x  x0 )
y  y ( x)  f ( x)  y0  y1
Sol: ( x0  x1 ) ( x1  x0 )
5. What is the Lagrange’s formula to find ‘y’ if three sets of values
.in
( x0 , y0 ),( x1 , y1 ) and ( x2 , y2 ) are given.
( x  x1 )( x  x2 ) ( x  x0 )( x  x2 ) ( x  x0 )( x  x1 )
ing
y y0  y1  y2
Sol: ( x0  x1 )( x0  x2 ) ( x1  x0 )( x1  x2 ) ( x2  x0 )( x2  x1 )
6. Find the second degree polynomial fitting the following data:
eer
x 1 2 4
Sol: y 4 5 13 Here x0 =
1, x1 = 2, x2 =4
gin
y0 = 4, y2 = 5, y3 = 13
By Lagrange’s formula for three points is
( x  x1 )( x  x2 ) ( x  x0 )( x  x2 ) ( x  x0 )( x  x1 )
y y0  y1  y2
En
( x0  x1 )( x0  x2 ) ( x1  x0 )( x1  x2 ) ( x2  x0 )( x2  x1 )
( x  2)( x  4) ( x 1)( x  4) ( x 1)( x  2)
y (4)  (5)  (13)
arn
(1  2)(1  4) (2 1)(2  4) (4 1)(4  2)

( x 2  6 x  8) ( x 2  5 x  4) ( x 2  3 x  2)
y (4)  (5)  (13)
Le
(1)(3) (1)(2) (3)(2)

( x 2  6 x  8) ( x 2  5 x  4) ( x 2  3 x  2)
 (4)  (5)  (13)
w.
3 2 6
1
 8 x 2  48 x  64  15 x 2  75 x  60  13 x 2  39 x  26 
6 
ww
1 2
 6 x  12 x  30 
6 
 y  f ( x)  x 2  2 x  5
Part B:
Eg.1. Using Lagrange interpolation formula, find f(4) given that
f (0)  2, f (1)  3, f (2)  12, f (15)  3578.
Sol:
Given x x0  0 x1  1 x2  2 x3  15
y0  2
y = f(x) y1  3 y2  12 y3  3587
Lagrange interpolation formula is

( x  x1 )( x  x2 )( x  x3 ) ( x  x0 )( x  x2 )( x  x3 )
y y0  y1
( x0  x1 )( x0  x2 )( x0  x3 ) ( x1  x0 )( x1  x2 )( x1  x3 )
( x  x0 )( x  x1 )( x  x3 ) ( x  x0 )( x  x1 )( x  x2 )
 y2  y3
( x2  x0 )( x2  x1 )( x2  x3 ) ( x3  x0 )( x3  x1 )( x3  x2 )
(4 1)(4  2)(4 15) (4  0)(4  2)(4 15)
f  4   2   3
(0 1)(0  2)(0 15) (1  0)(1  2)(1 15)
(4  0)(4 1)(4 15) (4  0)(4 1)(4  2)
 12    3587 
(2  0)(2 1)(2 15) (15  0)(15 1)(15  2)
(3)(2)( 11) (4)(2)( 11)
  2   3
( 1)(  2)( 15) (1)(1)( 14)
(4)(3)( 11) (4)(3)( 2)
 12    3587 
(2 )(1)( 13) (15)(14)(13)
.in
 77.99  78
CW.2. Find polynomial f(x) by using Lagrange formula from the given
ing
data and find f(8).
x 3 7 9 10
f(x) 168 120 72 63
eer
Sol: Lagrange polynomial f(x) = x3 – 21 x2 + 119 x – 27.
y(x = 8) or y(x = 8) or f(x = 8) = 93
HW.3. Use Lagrange’s formula to fit a polynomial to the data
gin
x –1 0 2 3
y = f(x) – 8 3 1 12
and hence find y(1).
En
Sol: Lagrange polynomial f(x) = 2 x3 – 6 x2 + 3 x + 3.

y(x = 1) = y(x = 1) = f(x = 1) = 2.
arn
Eg.4. Using Lagrange’s formula, prove that

y1  y3  0.3( y5  y3 )  0.2( y3  y5 ).
Sol: From the equation, the values of x are
Le
x0  5 x1  3 x2  3 x3  5
y0  y5 y1  y3 y2  y3 y3  y5
w.
The x values are not equally space, so use Lagrange’s

formula to find y = f(x). Lagrange’s formula for a set of 4 pair of
values is
ww
( x  x1 )( x  x2 )( x  x3 ) ( x  x0 )( x  x2 )( x  x3 )
y  yx  f  x   y0  y1
( x0  x1 )( x0  x2 )( x0  x3 ) ( x1  x0 )( x1  x2 )( x1  x3 )
( x  x0 )( x  x1 )( x  x3 ) ( x  x0 )( x  x1 )( x  x2 )
 y2  y3
( x2  x0 )( x2  x1 )( x2  x3 ) ( x3  x0 )( x3  x1 )( x3  x2 )
( x  3)( x  3)( x  5) ( x  5)( x  3)( x  5)
 y5  y3
(5  3)(5  3)(5  5) (3  5)(3  3)(3  5)
( x  5)( x  3)( x  5) ( x  5)( x  3)( x  3)
 y3  y5
(3  5)(3  3)(3  5) (5  5)(5  3)(5  3)

Put x  1, we get
(1  3)(1  3)(1  5) (1  5)(1  3)(1  5)
y1  y5  y3
(5  3)(5  3)(5  5) (3  5)(3  3)(3  5)
(1  5)(1  3)(1  5) (1  5)(1  3)(1  3)
 y3  y5
(3  5)(3  3)(3  5) (5  5)(5  3)(5  3)
 0.2 y5  0.5 y3  y3  0.3 y5
 0.2 y5  0.2 y3  0.3 y3  y3  0.3 y5
y1  y3  0.3( y5  y3 )  0.2( y3  y5 )
CW.5. Find the age corresponding to the annuity value 13.6 from the
given table
Age (x) 30 35 40 45 50
Annuity value (y) 15.9 14.9 14.1 13.3 12.5
.in
Sol: x13.6 or (y = 13.6) = 43
HW.6. Find x for which y = 7, given
ing
x 1 3 4
y 4 12 19
Sol: x(y = 7) = 1.8572
eer
AU1. Use Lagrange formula to calculate f(3) from the following table.
x 0 1 2 4 5 6
gin
f(x) 1 14 15 5 6 19
(Anna Univ. Nov./Dec., 2007(EEE)) Ans: f(3) = 10
AU2. Find the Lagrange’s polynomial of degree 3 to fit the data :
En
y(0) = −12, y(1) = 0, y(3) = 6 and y(4) = 12. Hence find y(2).
(M/J, 2007) Ans: f(x) = x3 − 7x2 + 18x − 12; y(2) = 4
arn
AU3. From the given table, the values of y are consecutive terms of a
series of which 23.6 is the 6th term. Find the first and tenth terms of
the series.
Le
x 3 4 5 6 4 5 6
y 4.8 8.4 14.5 23.6 36.2 52.8 73.9
w.
1
(N/D, 2007) Ans: y(x) = 12 [x3 + 3x2 − 14.8x + 48];
ww
y(1) = 3.1; y(10) = 100

AU4. Find the missing term in the following table using Lagrange’s
interpolation.
x 0 1 2 3 4
y 1 3 9 – 81
Ans: 31
AU5. Find the value of x corresponding to y = 100 from the table.
x 3 5 7 9 11
y 6 24 58 108 174
Ans: 8.656

Topic 2 : Divided Difference (Equally and unequally spaced intervals)

First divided difference for arguments x0 , x1 :
f ( x1 )  f ( x0 )
 f ( x0 , x1 )   f ( x0 )  [ x0 , x1 ](or )[ x1 , x0 ]   f ( x1 )
x1  x0 x1 x0
First divided difference for arguments x1 , x2 :

f ( x2 )  f ( x1 )
 f ( x1 , x2 )   f ( x1 )  [ x1 , x2 ](or )[ x2 , x1 ]   f ( x2 )
x2  x1 x2 x1
Second divided difference for arguments x0 , x1 , x2 :

 f ( x1 )   f ( x0 )
x2 x1
 f ( x0 , x1 , x2 )   2 f ( x0 )  [ x0 , x1 , x2 ]
.in
x2  x0 x2 x1
Second divided difference for arguments x1 , x2 , x3 :
ing
 f ( x2 )   f ( x1 )
x3 x2
 f ( x1 , x2 , x3 )   2 f ( x1 )  [ x1 , x2 , x3 ]
x3  x1
eer
x3 x2
Third divided difference for arguments x0 , x1 , x2 , x3 :

 2 f ( x1 )   2 f ( x0 )
gin
x3 x2 x2 x1
 f ( x0 , x1 , x2 , x3 )  3 f ( x0 )  [ x0 , x1 , x2 , x3 ]
x3  x0 x3 x2 x1
En
Third divided difference for arguments x1 , x2 , x3 , x4 :

 2 f ( x2 )   2 f ( x1 )
x4 x3 x3 x2
 f ( x1 , x2 , x3 , x4 )  3 f ( x1 )  [ x1 , x2 , x3 , x4 ]
arn
x4  x1 x4 x3 x2
Properties of divided differences:
Le
1. The divided difference are symmetric functions of their arguments.

 f ( x)   f ( y )
For example, y x
w.
 2 f ( x)   2 f ( y )   2 f ( z )
yz xz xy
ww
2. The nth divided differences of a polynomial of degree n are

constants.
3. The divided difference operator(  ) is a linear operator.
[ f ( x)  g ( x)]  [ f ( x)]  [ g ( x)]
and [cf ( x)]  c[ f ( x)] , c is constant.
Part A:
 2 x3  x  y  z
7. Prove that yz .
3
Sol: Given the function f ( x)  x and the arguments are x,y,z.

f ( y )  f ( x) y  x3 ( y  x)( x 2  xy  y 2 )
3
 f ( x)     x 2  xy  y 2
y yx yx yx
 f ( y )  y 2  yz  z 2
Similarly, z
 f ( y )   f ( x)
2 z y y 2  yz  z 2  ( x 2  xy  y 2 )
Now  f ( x)  
yz zx zx
2 2
z  x  yz  xy ( z  x)( z  x)  y ( z  x)
   x yz
zx zx
1 1
 3    
8. Show that bcd a
  abcd .
.in
1 1
f ( x)  , f (a) 
Sol: If x a
ing
1 1

1 f (b)  f (a ) b a 1
f (a, b)     
b a ba ba ab
eer
1 1 1 1
f (b, c)     , f (c, d )    
Similarly, c b bc d c cd
gin
1 1 1 1
    
21 c b b b bc ab a  c 1
f (a, b, c)      
bc a ca ca abc(c  a ) abc
En
1 1
f (b, c, d )   2 
Similarly, cd b bcd
arn
1 1 1 1
 2   2 
1
f (a, b, c, d )   3  cd b bc a  bcd abc
 bcd a d a d a
Le
ad 1
 
abc(d  a ) abcd
w.
Part B:
Eg.7. Construct the divided difference table for the following data and find
ww
the value f(2).

x 4 5 7 10 11 12
y = f(x) 50 102 296 800 1010 1224
Sol: Newton’s divided difference formula is
f  x   f  x0    x  x0  f  x0 , x1    x  x0   x  x1  f  x0 , x1 , x2 
  x  x0   x  x1  x  x2  f  x0 , x1 , x2 , x3  

Newton’s divided difference table is

f ( x)  2 f ( x) 3 f ( x)  4 f ( x) 5 f ( x)
x f(x)     
f  x0 , x1  f  x0 , x1 , x2  f  x0 ,x1 ,x2 ,x3  f  x0 ,x1 ,x2 ,x3 ,x4  f  x0 ,x1,x2 ,x3 ,x4 ,x5 
4 50
102  50
 52
54
9752
5 102  15
74
296102 14.215
 97  0.133
75 10  4
.in
16897 .617.133
7 296  14.2  .069
10  5 11  4
ing
800296 10.514.2 .158.069
 168  .617  .011
10  7 11  5 12  5
210168 1.7.617
eer
10 800  10.5  .158
11  7 12  5
1010800 210.5
 210  1.7
gin
11  10 12  7
214210
11 1010  2
12  10
En
12241010
 214
12  11
arn
12 1224
f  x   50   x  4   52    x  4   x  5  15    x  4   x  5  x  7   0.133
  x  4   x  5  x  7   x  10   0.069 
Le
  x  4   x  5  x  7   x  10   x  11  0.011
f  x  2   50   2  4   52    2  4   2  5 15    2  4   2  5  2  7  0.133
  2  4   2  5  2  7  2  10  0.069 
w.
  2  4   2  5  2  7  2  10  x  11 0.011
 49.19
ww
CW.8. If f(0) = 0, f(1) = 0, f(2) = –12, f(4) = 0, f(5) = 600, f(7) = 7308,
find a polynomial that satisfies this data using Newton’s devided difference
formula. Hence find f(6), f(–1).
3 2
Sol: f(x) = x( x  1)[ x  2 x  13x  20]
f(6) = 2580 , f(–1) = 60

HW.9. Find the third divided difference with arguments 2,4,9,10 of the
function f(x) = x3 – 2x.
Sol: Form divided difference table.
AU6. If f(0) = f(1) = 0, f(2) = −12, f(4) = 0, f(5) = 600 and f(7) = 7308, find
a polynomial that satisfies this data using Newton’s divided
difference interpolation formula. Hence, find f(6)
(Anna Univ. May./June, 2007) Ans: f(6) = 2580
AU7. Given the values
x 5 7 11 13 17
f(x) 150 392 1452 2366 5202
Evaluate f(9) using Newton’s divided difference formula.
(N/D, 2007) Ans: f(x) = x3 − x2 + 24x − 70; f(9) = 794
.in
AU8. Using Newton’s divided difference interpolation, find the
polynomial of the given data
ing
x –1 0 1 3
f(x) 2 1 0 –1
1
eer
(Anna Univ. Nov./Dec., 2007) Ans: f(x) = 24 [x3 − 25x + 24]
gin
En
arn
Le
w.
ww

Topic 3 : Interpolating with a cubic spline

The cubic spline interpolation formula is
1
S ( x)  y ( x)  y  [( xi  x)3 M i 1  ( x  xi 1 )3 M i ]
6h
1 h2 1 h2
 ( xi  x)[ yi 1  M i 1 ]  ( x  xi 1 )[ yi  M i ]
h 6 h 6
6
M i 1  4M i  M i 1  2 [ yi 1  2 yi  yi 1 ]
where h
n = number of data
i = number of intervals [ i.e., i = 1,2,3,…, (n – 1)]
h = length f interval = interval length.
''
Note: If M i and yi values are not given, then assume
.in
M 0  M n  0 [or y ''  y ''  0] M , M ,..., M n 1
0 n , and find 1 2
ing
in 1st interval,2nd interval,…, (n–1)th interval value.
Part A:
9. What is a cubic spline?
eer
Sol: A cubic spline which has continuous slope and curvature
is called a cubic spline.
10. What is a natural cubic spline?
gin
Sol: A cubic spline fitted to the given data such that the end
cubics approach linearity at their extremities is called a
natural cubic spline.
En
11. State the conditions required for a natural cubic spline.

Sol: A cubic spline g(x) fits to each of the points is continuous
arn
and is continuous in slope and curvature such that

M 0  S0  g0' ' ( x0 )  0 and M n  Sn  g n' ' 1 ( xn )  0 is
called a natural cubic spline. Let us assume that
Le
( xi , yi ), i  0,1, 2,..., n are data points.

12. What are the advantages of cubic spline fitting?
w.
Sol: Cubic spline provide better approximation to the behavior

of functions that have abrupt local changes. Further,
ww
spline perform better than higher order polynomial

approximation.
13. Write the end conditions on M i (x) in natural cubic spline.
Sol: M 0 ( x)  0, M n ( x)  0.
14. Write the relation between the second derivatives M i (x) in cubic
splines with equal mesh spacing.
6
M i 1  4M i  M i 1  2 [ yi 1  2 yi  yi 1 ]
Sol: h ,i = 1,2,…,n – 1.
Or
6
M i 1  4 M i  M i 1  [ fi 1  2 fi  fi 1 ]
h2 ,i = 1,2,…,n – 1.

Part B:
Eg.10. Find the cubic spline approximation for the function f(x) given by
the data:
x 0 1 2 3
y = f(x) 1 2 33 244
with M 0  0  M 3. Hence estimate the value f(1.5),f(2.5).
{AU2010}
Sol: We know that cubic spline interpolation formula for
xi 1  x  xi , i  1, 2,3. is
Si  x   y  x   y 
1 
  xi  x 3 M i1   x  xi1 3 M i 
.in
6h 
1  h2 
  xi  x   yi 1  M i 1 
ing
h  6 
1  h2 
  x  xi 1   yi  Mi   (1)
h  6 
eer
6
where Mi 1 +4Mi  Mi 1 = 2  yi 1  2 yi  yi 1   (2)
h
gin
h  number of data  4
i  number of intervals  3 i.e., i  1, 2,3.
h  length of inteval  1
En
Given M 0  M 2  0, so find M1 , M 2 .
Suppose M i or yi" values are not given, then assume M 0  M 3  0 
arn
 and find M1 , M 2 . 
 
To find M1 , M 2 :
Le
6
When i  1, (2)  M 0  4 M1  M 2   y0  2 y1  y2 
12
 0  4 M1  M 2  6 1  2  2   33
w.
 4 M1  M 2  180  (3)
6
When i  2, (2)  M1  4 M 2  M 3  2  y1  2 y2  y3 
ww
1
 M1  4 M 2  0  6  2  2  2   244
 4 M1  M 2  180  (4)
Solving  3 &  4  ,  3  4 M1  M 2  180
4   4   4 M1  16 M 2  4320
 15M 2  4140
 M 2  276
 3  4 M1  180  276  M1  24

11 .
To find Cubic spline:

When i  1,
Cubic spline in xi 1  x  xi
i.e., x0  x  x1
i.e., 0  x  1
i.e., Cubic spline in 0  x  1 is
1 
y1  x   S1  x    x1  x 3 M 0   x  x0 3 M1 
6 1 
1  1  1  1 
  x1  x   y0  M 0    x  x0   y1  M1 
1  6  1  6 
1 3 3 
   
  1  x 0  x  0 24    
6
 1  x  1  0   x  0   2   24 
.in
 4 x3  1  x   6 x
 4 x3  5 x  1
ing
When i  2,
i.e., x1  x  x2
eer
i.e., 1  x  2
1 
y2  x   S 2  x     x2  x 3 M1   x  x1 3 M 2 
gin
61 
1  1  1  1 
  x2  x   y1  M1    x  x1   y2  M 2 
1  6  1  6 
En
1
  2  x   24    x  1  276  
3 3
6
 1   1 
  2  x   2   24     x  1 33   276  
arn
 6   6 
3 3
 4  2  x   46  x  1  6  2  x   13  x  1
 50 x3  162 x 2  162 x  53
Le
When i  3,
w.
i.e., x2  x  x3
i.e., 2  x  3
ww
1 
y3  x   S3  x    x3  x 3 M 2   x  x2 3 M 3 
6 1 
1  1  1  1 
  x3  x   y2  M 2    x  x2   y3  M 3 
1  6  1  6 
1 3 
  3  x   276   0 
6
 1 
  3  x  33   276     x  2   244  0
 6 
 46  27  x3  9 x 2  27 x   13  3  x   244 x  488
 46 x3  414 x 2  985 x  715

 Cubic spline is
 S1 ( x)  y1 ( x)  4 x3  5 x  1, 0  x 1

S ( x)   S2 ( x)  y2 ( x)  50 x3  162 x 2  167 x  53, 1  x  2
 S3 ( x)  y3 ( x)  46 x3  414 x 2  985 x  715, 2  x  3

When x  1.5,
3 2
y2 ( x  1.5)  S2 ( x  1.5)  50 1.5   162 1.5   167 1.5   53
 1.75,
When x  2.5,
3 2
y3 ( x  2.5)  S3 ( x  2.5)  46  2.5   414  2.5   985  2.5   715
 121.25
CW.11. From the following table .
.in
x 1 2 3
y = f(x) –8 –1 18
ing
' '
Find cubic spline and compute y(1.5), y (1), y (2.5) and y (3).
 S ( x)  y1 ( x)  3( x  1)3  4 x  12, 1  x  2
S ( x)   1
eer
3
Sol:  S2 ( x)  y2 ( x)  3(3  x)  22 x  48, 2  x  3
&
gin
45
y ( x  1.5)  S1 ( x  1.5)   , y ' ( x  1)  S1' ( x  1)  4
8
y ( x  2.5)  S2 ( x  2.5)  7.375, y ' ( x  3)  S2' ( x  3)  22.
En
HW.12. Fit a natural cubic spline for the following data: {AU2008}
x 0 1 2 3
arn
y = f(x) 1 4 0 –2
Sol: Assume M 0  0  M 3.
Le
 S1 ( x)  y1 ( x)  2 x3  5 x  1, [0,1]

S ( x)   S2 ( x)  y2 ( x)  3 x3  15 x 2  20 x  4, [1,2]
w.
 S3 ( x)  y3 ( x)   x3  9 x 2  28 x  28, [2,3]

ww
AU9. Find the cubic Spline interpolation. (N/D, 2007)

X 1 2 3 4 5
f(x) 1 0 1 0 1
Sol: Assume M 0  0  M 4 .
 S1 ( x)  y1 ( x )  2  x, 1 x  2
 1 3 2
 S2 ( x)  y2 ( x )  [ 5 x  45 x  123 x  106], 2 x3
 7
f ( x)  S ( x)   1 3 2
S ( x)  y3 ( x )  [6 x  72 x  275 x  332], 3 x4
 3 7
 1
 S4 ( x)  y4 ( x )  [ 5 x3  75 x 2  363 x  772], 4 x5
 7

13 .
AU10.Given the following table, find f(2.5) using cubic spline functions :
x 1 2 3 4
(Anna Univ. f(x) 0.5 0.3333 0.25 0.2 May./June,
2007) Ans: S2(2.5) = 0.2829
AU11.Fit the st.line for the data. (Anna Univ. May/June, 2007)
x 0 1 2 3
f(x) 1 2 9 28
 4 3 4
 y1 ( x)  5 x  5 x  1, 0  x 1
 1
f ( x)   y2 ( x)  [10 x3  18 x 2  19 x  1, 1 x  2
5
.in
 102 2 333 159
 y3 ( x)  2 x3  x  x , 2 x3
 5 5 5
ing
Topic 4 : Newton’s forward and backward difference formulas
*[Uniform (or) equal intervals only] *
eer
Newton’s forward interpolation difference formula:
[if y(required x near to x0 ) = ? and use  ]
gin
y ( x)  f ( x)  f ( x0  uh)
u u (u  1) 2 u (u  1)(u  2) 3
= y0  y0   y0   y0  ...
1! 2! 3!
En
x  x0
u ,
where h h = length of interval
arn
Newton’s backward interpolation difference formula:

[if y(required x near to xn ) = ? and use  ]
y ( x)  f ( x)  f ( xn  vh)
Le
v v(v  1) 2 v(v  1)(v  2) 3

 yn  yn   yn   yn  ...
1! 2! 3!
x  xn
w.
v ,
ww
Part A:
15. What advantage has Lagrange’s formula over Newton?
Sol: The forward and backward interpolation formulae of
Newton can be used only when the values of the independent
variable x are equally spaced can also be used when the
differences of the dependent variable y become smaller
ultimately. But Lagrange’s interpolation formula can be used
whether the values of x, the independent variable are equally
spaced or not and whether the difference of y become smaller
or not.

16. Derive Newton’s forward difference formula by using operator
method. [or] Derive Gregory – Newton forward difference
interpolation formula.
.in
ing
eer
gin
En
arn
Le
w.
ww

15 .
Sol: Pn  x   Pn  x0  uh   E u Pn  x0   E u y0
 1    y0
u
u u (u  1) 2 u (u  1)(u  2) 3
= y0  y0   y0   y0  ...
1! 2! 3!
x  x0
u
where h
17. Derive Newton’s backward difference formula by using
operator method.
Sol: Pn  x   Pn  x0  vh   E v Pn  xn   E v yn
v 1
 1    yn where E  1   
.in
 v(v  1) 2 v(v  1)(v  2) 3 
 1  v      ...  yn
ing
 2! 3! 
v v(v  1) 2 v(v  1)(v  2) 3
 yn  y n   yn   yn  ...
1! 2! 3!
eer
x  xn
v ,
where h
gin
18. When will we use Newton’s forward interpolation formula?

Sol: The formula is used to interpolate the values of y near
the beginning of the table values and also for
En
extrapolating the values of y short distance ahead (to

the left) of y0.
19. When Newton’s backward interpolation formula is used?
arn
Sol: The formula is used mainly to interpolate the values of

y near the end of a set of tabular values and also for
extrapolating the values of y short distance ahead (to
Le
the right) of yn.

Part B:
w.
Eg.13.Using New ton’s forward interpolation formula, find the cubic

polynomial which takes the following values
x 0 1 2 3
ww
y = f(x) 1 2 1 10
Evaluate f(4). {AU 2000, 2009}
Sol: WKT, Newton forward formula to find the polynomial in x.
There are only 4 data given.Hence the polynomial will be degree 3.
Newton’s forward formula is
y ( x)  f ( x)  f ( x0  uh)
u u (u  1) 2 u (u  1)(u  2) 3
= y0  y0   y0   y0  ...
1! 2! 3!
x  x0
u ,

Newton’s divided difference table

x y  f ( x) y 2 y 3 y
x0  0 y0  1
2 1  1
x1  1 y1  2 1  1  2
1 2  1 10   2   12
x2  2 y2  1 9   1  10
10  1  9
x3  3 y3  10
x   x  x  1   x  x  1 x  2   
.in
f  x  1 1  2  12
1! 2! 3!
 2 x3  7 x 2  6 x  1
ing
When x  2,
f (4)  41
CW.14. The population of a city in a census takes once in 10 years is
eer
given below. Estimate the population in the year 1955.
Year 1951 1961 1971 1981
gin
Population in lakhs 35 42 58 84
Sol: y ( x  1955)  f ( x  1955)  36.784
HW.15. From the table given below find sin 52 by using Newton’s
En
forward interpolation formula.

x 45  50  55  60 
arn
y = sin x 0.7071 0.7660 0.8192 0.8660

Sol: y ( x  52)  sin 52  0.788 approximately.
Eg.16.From the data given below find the number of students whose
Le
weight is between 60 and 70.

Weight in lbs 0–40 40–60 60–80 80–100 100–120
w.
Number of students 250 120 100 70 50

Sol: Let weight be denoted by x and
number of students be denoted by y, y = f(x).
ww
Use Newton's forward formula to find y where x lies

between 60 – 70.
Newton's forward formula is
y ( x)  f ( x)  f ( x0  uh)
u u (u  1) 2 u (u  1)(u  2) 3
= y0  y0   y0   y0  ...
1! 2! 3!
We rewrite the table as cumulative table showing the number of
students less than x lbs.
x Below 40 Below 60 Below 80 Below 100 Below 120
y 250 370 470 540 590
Newton's forward difference table:

17 .
x y  f ( x) y 2 y 3 y 4 y
Below 40 250
120
Below 60 370 –20
100 –10
Below 80 470 –30 20
70 10
Below 100 540 –20
50
Below 120 590
x  x0 x  40 70  40
u  x  70, u   1.5
20
.in
Here h 20 , When
1.5 1.5(1.5  1) 1.5(1.5  1)(1.5  2)
y ( x  70)  250  120    20    10 
1! 2! 3!
ing
1.5(1.5  1)(1.5  2)(1.5  3)
  20 
4!
 423.59  424
eer
 Number of students whose weight is 70  424
 Number of students whose weight is between 60  70  424  370  54.
CW17. Use Newton’s backward difference formula to construct as
gin
interpolating polynomial of degree 3 for the data.

f(– 0.75) = – 0.07181250, f(– 0.5) = –0.024750,
1
En
f ( ).
f(– 0.25) = 0.33493750, f(0) = 1.10100. Find 3
3 2
y ( x)  x  4.001x  4.002 x  1.101 f ( ) 1
arn
Sol: , 3 = 0.174518518
HW.21. From the following data, find  at x  43 and x = 84.
Le
x 40 50 60 70 80 90
 184 204 226 250 276 304
Also express  in terms of x.
w.
Sol:  ( x  43)  189.79 (by Newton;s forward formula)

 ( x  84)  286.96 (by Newton;s backward formula)
ww
 ( x)  0.01x 2  1.1x  124 (by Newton;s forward formula)

AU12. Find a polynomial of degree two for the data by Newton’s forward
difference method :
x 0 1 2 3 4 5 6 7
f(x) 1 2 4 7 11 16 22 29
1 2
[ x  x  2]
(Anna Univ. May./June, 2007) Ans: y(x) = 2

Unit II Assignment problems:

1. Find the missing term in the following table using Lagrange’s
interpolation.
x 0 1 2 3 4
y 1 3 9 – 81
Ans: 31
 f ( )  0.3887 
d
f ( )  
0 sin 2 
1
2. Find the value of given where 2
 21 23 25
using the table f   0.3706 0.4068 0.4433
.in
Sol:  = 22.0020
3. Using Newton’s divided difference formula find f(x) and f(6) from the
ing
following data : (M/J, 2007(EEE))
x 1 2 7 8
eer
f(x) 1 5 5 4
1
Ans:f(x) = 42 [3x3 − 58x2 + 321x − 224]; f(6) = 6.2381
gin
4. Using Newton’s divided difference formula, find the value of f(8)

and from the following table:
x 4 5 7 10 11 13
En
f(x) 48 100 294 900 1210 2028

(Anna Univ. Nov./Dec., 2007(EEE)) Ans: f(8) = 448
arn

(0, 0), ( ,1) & ( , 0)
5. Given the points 2 satisfying the function
y  sin x(0  x   ) 
Le
y( )
, determine the value of 6 using the cubic
spline approximation. [ Assume M 0  0  M 2. ]
w.
2  2 3 3        
S ( x)   x  x  , 0,  y    S    0.4815
  2 2   2 &   6 6
ww
Sol:
6. Fit the cubic spline for the following data:
x 0 1 2 3
y = f(x) 1 2 9 28
Sol: Assume M 0  0  M 3.
 4 3 x
 S1 ( x)  y1 ( x)  5 x  5  1, [0,1]
 1
S ( x)   S2 ( x)  y2 ( x)  (10 x3  18 x 2  19 x  1, [1,2]
 5
1
 S3 ( x)  y3 ( x)  (14 x3  126 x 2  269 x  191, [2,3]
 5

19 .
7. From the following table, find the value of (tan45o15’) by Newton’s

forward interpolation formula.
xo 45 46 47 48 49 50
tan x o 1.00000 1.03553 1.07237 1.11061 1.15037 1.19175
(M/J, 2007(EEE)) Ans: f(45o15’) = 1.00876
8. Given:
x 1 2 3 4 5 6 7 8
f(x) 1 8 27 64 125 216 343 512
Estimate f(7.5). Use Newton’s formula.
(Anna Univ. Nov./Dec., 2007(EEE)) Ans: f(7.5) = 421.87
9. The hourly declination of the moon on a day is given below. Find the
.in
h m s h
declination at 3 35 15 and 5 .
ing
Hour 0 1 2 3 4
Declination 82953.7 81819.4 8643.5 7556.1 74327.2
Ans:    
y 3h35m15s  74815, y 5h  1.00876
eer
gin
En
arn
Le
w.
ww

.in
ing
eer
gin
En
arn
Le
w.
ww

21 .
.in
ing
eer
gin
En
arn
Le
w.
ww

Unit III NUMERICAL DIFFERENTIATION AND INTEGRATION

Differentiation using interpolation formulae –Numerical integration by
trapezoidal and Simpson’s 1/3 and 3/8 rules – Romberg’s method – Two
and Three point Gaussian quadrature formulae – Double integrals using
trapezoidal and Simpsons’s rules.
Numerical Differentiation Numerical Integration

1. Newton’s forward difference 6. Trapezoidal rule [n =1 in
.in
formula to compute derivatives Quadrature formula]
(Equal interval)
ing
2. Newton’s backward difference 7. Simpson’s one third rule
formula to compute derivatives [n = 2 in Quadrature formula]
(Equal interval)
eer
3. Lagrange’s Interpolation 8. Simpson’s three eighth rule
formula [n = 3 in Quadrature formula]
(Equal or unequal intervals)
gin
4. Newton’s divided difference 9. Romberg’s method

Interpolation formula
(Equal or unequal intervals)
En
5. Maxima and minima of a 10. Two point Gaussian’s

tabulated function quadrature formula
11. Three point Gaussian’s
arn
quadrature formula
12. Double integrals by
Trapezoidal rule
Le
13. Double integrals by Simpson’s

rule
w.
NUMERICAL DIFFERENTIATION:
ww
Given y  y  x   f  x  [in a table]

dy
 y  x   f   x 
dx is the first numerical derivative
d2y
 y   x   f   x 
dx 2 is the second numerical derivative
n
d y
n
 y1⋯n  x   f 1⋯n  x 
dx is the nth numerical derivative

2 Unit III – Numerical Differentiation and Integration
Topic 1: Newton’s forward difference formula to compute derivative

WKT, Newton’s forward difference interpolation formula is
u u  u  1 2 u  u  1 u  2  3
y  x   f  x0  uh   y0  y0   y0   y0
1! 2! 3!
u  u  1 u  2  u  3 4
  y0  ...
4!
u u2  u 2 u 3  3u 2  2u 3
 y0  y0   y0   y0
1! 2! 3!
u 4  6u 3  11u 2  6u 4
  y0  ...
.in
4!
x  x0
u
ing
where h
First derivative :
dy dy du dy  1  1 dy
eer
   
dx du dx du  h  h du
dy 1   2u  1  2  3u 2  6u  2  3
gin
i.e.,   y0     y0     y0
dx h   2   6 
 4u 3  18u 2  22u  6  4 
En
   y0  ...
 24  
 
arn
 dy  1 1 1 1 
 dx  at   y0   2 y0  3 y0   4 y0  ......
  x  x0  h 2 3 4 
u  0 
Le
Second derivative :
d 2 y d  dy  d  dy du  d  1 dy 
      
w.
dx 2 dx  dx  dx  du dx  dx  h du 
1 d  1 dy  1 d 2 y
ww
 
h du  h du  h 2 du 2
1   12u 2  36u  22  4 
 2   2 y0   u  1 3 y0     y0  ......
h   24  
 
d2y 1  2 22 
 2  2 
 y0  3 y0   4 y0  .....
 dx  atx  x0  h  24 
u  0 

Third derivative :
d3y 1  3  24u  36  4 
3
 3 
 y0     y0  .....
dx h   24  
d3y  1  3 36 
 3  3 
 y0   4 y0  .....
 dx  at x  x0  h  24 
u  0 
Topic 2: Newton’s backward difference formula to compute derivatives

v  v  1 2 v  v  1 v  2  3
.in
v
y  x   f  xn  vh   yn  yn   yn   yn
1! 2! 3!
v  v  1 v  2  v  3 4
ing
  yn  ...
4!
v  v2  v  2  v3  3v 2  2v  3
eer
 yn  yn     yn     yn
1!  2!   3!
   
 v  6v  11v  6v  4
4 3 2
gin
   yn  .....
 4!
 
First derivative :
En
dy dy dv dy  1  1 dy
    a
dx dv dx dv  h  h dv
arn
1  2v  1  2  3v 2  6v  2  3
   yn     yn     yn
h   2   3! 
Le
 4v3  18v 2  22v  6  4 

   yn  ...
 
w.
 4!  
 dy  1 1 2 6 
 yn   2 yn  3 yn   4 yn  ...
ww
 dx  at x  x0  h 2 3! 4! 
u 0 
Second derivative :
d2y 1  2  6v  6  3  12v 2  36v  22  4 
   y     y     y  ...
dx 2 h 2 
n
 3! 
n  4!  n

 
d2y 1  2 6 22 
 2  2 
 yn  3 yn   4 yn  ...
 dx  atx  x0  h  3! 4! 
u  0 

Third derivative :
d3y 1 6 3  24v  36  4 
3
 3  3!
 yn     yn  .....
dx h   4!  
d y
3
1 6 3 36 
 3  3  3!
 yn   4 yn  .....
 dx  atx  x0  h  4! 
u  0 
Topic 3: Maxima and Minima of a tabulated function

(If the intervals are same)
.in
u u  u  1 2 u  u  1 u  2  3
y  y0  y0   y0   y0  ...
ing
1! 2! 3!
First derivative :
dy 1  2u  1 2 3u 2  6u  2 3 
  y0   y0   y0  ...  1
eer
dx h  2! 3! 
substitute h, y0 , 3 y0 ,  4 y0 , ... gives the equation
gin
dy u   2

dx an equation in
En
For maxima or minima is obtained by the equation by

dy
 0 (or) y   x  =0
arn
dx
Put RHS of (1) =0, find u  u1 , u2 , ...
d2y
Le
2
Find dx from (2),

 d2y 
w.
 2  
 dx at u
Suppose 1 ve
ww
 u is maximum point with y  u   maximum value of y

1 1
 d2y 
Suppose  2   + ve  u2 is minimum point with
 dx 
 at u2
y  u2  = minimum value of y

Examples on Numerical Differentiation

Eg (1) : The following data gives the velocity of a particle for 20 seconds at
an interval of 5 seconds.
Time
0 5 10 15 20
(sec)
Velocity
0 3 14 69 228
(m/sec)
Find (a) Initial acceleration using the entire data
(b) Final acceleration
Solution : The difference table is
.in
Time Velocity y  x  2 y  x  3 y  x  4 y  x 
t=x v =y(x)
ing
0  x0  0  y  0  
3  y  0  
eer
5 3 8  2 y  0 
 
36  3 y  0  
gin
11  
 4 y  0 
 
En
10 14 44 24   or  
 4 
  yn 
arn
55 
60 3 yn 
 
Le
15 69 104  2 yn
159  yn 
w.
20 228  yn 
ww
Here x0  0 , h = interval length = 5

dv
(a) WKT, acceleration = dt = rate of change of velocity
 dv 
 
To find initial acceleration, put  dt t 0
ie., initial acceleration exists at t  0  x0 [which is nearer to beginning of
the table], so we use Newton’s forward difference formula for first
derivative.
 Newton’s forward difference interpolation formula is

u u  u  1 2
y  x   f  x0  uh   y0  y0   y0
1! 2!
u  u  1 u  2  3
  y0  ...
3!
Now, Newton’s forward difference formula for first derivative
at x  x0  0  u  0 ,
 dv  1 1 1 1 
y   x0       y0   2 y0  3 y0   4 y0  ...
ie., dt
 t  0 h  2 3 4 
[Here h = 5]
1 1 1 1 
 3   8    36    24    1
.in
5 2 3 4 
 y  0  1
ing
 2
Initial acceleration (acceleration when t = 0) is 1 m/sec .
(b) Final acceleration exists at t = 20 = x4 [Nearer to ending of table], so
eer
use Newton’s backward difference interpolation formula for first derivative.
WKT, Newton’s backward difference interpolation formula is
v  v  1 2 v  v  1 v  2  3
gin
v
y  x   f  xn  vh   yn  yn   yn   yn
1! 2! 3!
v  v  1 v  2  v  3 4
En
  yn  ...
4!
1 1 2 6 
y   xn   yn   2 yn  3 yn   4 yn  ...
arn
h 2 3! 4! 
1 1 2 6 
 159  104    60    24  
Le
5 2 6 24 
1
  237   47.2 m / sec 2
w.
5
Eg (2) : Consider the following table of data :
ww
x 0.2 0.4 0.6 0.8 1.0

f(x) 0.9798652 0.9177710 0.8080348 0.636093 0.3843735
Find (1) f   0.25  using Newton’s forward difference approximation
(2) f   0.95  using Newton’s backward difference approximation
Solution : Here h = 0.2
x  x0 x  0.2 0.25  0.2
u    0.25
(1) h 0.2 0.2
 f   0.25   0.25828

x  xn 0.95  1
u   0.25
(2) h 0.2
f   0.95   1.367948
Eg (3) : Find the value of f   8  , f   9  , maximum and minimum value

from the following data, using an approximate interpolation formula.
x 4 5 7 10 11
f (x) 48 100 294 900 1210
Solution : The values of x are unequally spaced. To find f (x), we use
Newton’s divided difference formula (or) Lagrange formula.
WKT, Newton’s Divided difference formula is
f  x   f  x0    x  x0  f  x0 , x1    x  x0   x  x1  f  x0 , x1 , x2 
.in
  x  x0   x  x1  x  x2  f  x0 , x1 , x2 , x3   ...
ing
x f(x) f  x   2 f  x  3 f  x   4 f  x 
4 48
eer
52
5 100 15
97 1
gin
7 294 21 0
202 1
10 900 27
En
310
11 1210
arn
 f  x   x3  x 2
f   x   3x 2  2 x , f   8   176
f   x   6 x  2 f   9   52
Le
,
To find maximum and minimum
w.
Put f   x   0  3x 2  2 x  0  x  0, x2
3
Now, f   x  0   6  0   2  2  0
ww
 x = 0 is a maximum pt. & maximum value is f (x= 0) =0

f   x  2 3  6  2 3  2  2  0
 x  2 3 is a minimum point & minimum value is f  x  2 3  4 / 27
HW 4 : Evaluate y  and y  at x = 2 given

x 0 1 3 6
y 18 10 – 18 40

 70 2 15 187 22 
Solution :y  x   x  9 x  9 x  18, y   x  2    9 , y   x  2    9 
3
 
HW 5 : Find the value of cos (1.747) using the values given in the
table below :
x 1.70 1.74 1.78 1.82 1.86
sin x 0.9916 0.9857 0.9781 0.9691 0.9584
[Solution : 0.175]
HW 6 : Find sec 31˚ from the following data :
θ 31 32 33 34
tan θ 0.6008 0.6249 0.6494 0.6745
Solution : sec 2 31  1.3835  sec 31  1.174 
 
.in
NUMERICAL INTEGRATION
Newton-Cote’s quadrature formula is

a  nh  n n  2 n  3  ing
2 n  n  2
2 
eer
 f  x  dx  nh  1      3
 ... f a
a  2 12 24 
gin
Topic 6: Trapezoidal rule(n = 1 in quadrature formula)

h
 2  y0  yn   2  y1  y2  y3  ...  yn 1  
En
x0  nh 
x0 f  x  dx   or 
arn
h
  first term  last term   2  remaining terms  
 2  
Le
Topic 7: Simpson’s one third rule (Simpson’s 1/3 rule)

(n = 2 in quad. form.)

w.
 h  y  y   4  y  y  y  ...  2  y  y  y  ... 
3  0 n 1 3 5 2 4 6 
ww
x0  nh 
x0 f  x  dx   or 

 h  first term  last term  
 3  4  odd suffices   2  even suffices  
  
Topic 8: Simpson’s three eighth rule (Simpson’s 3/8 rule) (n=3 in q.f.)

  y0  yn  
 3h  
  2  y3  y6  ... 
8 
   1 3 y  y 2  y 4  y5  y7  ...  
x0  nh 
x0 f  x  dx   or 

  first term  last term  
 3h  2  suffices with a multiple of 3 
8  
  3  remaining terms  
  
 b 
 I =  f  x  dx 
.in
 
Topic 9: Romberg’s method for a given interval  a 
ba
ing
when h  , by trapezoidal rule, we get I1
2
ba
when h  , by trapezoidal rule, we get I 2
eer
4
ba
when h  , by trapezoidal rule, we get I 3
gin
8
I  I 
Romberg's formula for I1 & I 2  I RM1,2  I 2  2 1
3
En
 I3  I 2 
Romberg's formula for I 2 & I 3  I RM 2,3  I 3 
3
arn
If I RM1,2  I RM 2,3 , then we can equal I  I RM1,2  I RM 2,3

Check, use actual integration, we get I AI  I  I RM1,2  I RM 2,3
Le
Topic 10: Two point Gaussian quadrature formula :

(Guass two point formula)
w.
b
Given I   f  x  dx
a
ww
Case (i) If a  1, b  1, then

1  1   1 
I  f  x  dx  f    f  
1
 3   3
Case (ii) If a  0, b  1, then
1 1 1
I   f  x  dx  f  x  dx, if f  x  is an even function
0 2 1
Case (iii) If  a  1& b  1 , then

ba ba
x z  mz  c  dx  m dz
2 2
b 1 1
 I   f  x  dx   f  z  m dz  m 1 f  z  dz
a 1
  1   1 
 m f   f 
  3  3 
Topic 11: Three point Gaussian quadrature formula
1 5  3   3  8
(i) 1 f  x  dx  9 f     f     f  0 
.in
 5   5   9
1 1 1
0 f  x  dx  2 1 f  x  dx for even function f  x  
ing
(ii)
1  5   3  3  8 
  f     f     f  0  
eer
2  9   5  5   9 
(iii)  a  1 & b  1 , then
gin
ba ba
x z  mz  c  dx  m dz
2 2
 5   3  
En
b 1  3  8
 I   f  x  dx   f  z  m dz  m   f     f     f  0  
1
 9   5   5   9
a

arn
Topic 12: Double integration by trapezoidal rule

d b
c a f  x  dx dy
Le
Given
sum of values of f at the four corners   D  a, d  C  b, d 

w.
 
hk  sum of values of f at the nodes   k↕
I 2   
4  on the boundary except the corners  
ww
4 sum of the values at the interior nodes  A  a, c  B  b, c 

 

h
ba d c
where h  ,k
n m
where n is number of equal intervals in  a, b  .
where m is number of equal intervals in  c, d  .

11 .
Topic 13: Double integration by Simpson’s 1/3 rule

d b
Given c a f  x, y  dx dy
 
 
sum of values of f at the four corners  
 sum of values of f at the odd positions  
2   
 on the boundary except the corners  
 
4 sum of the values of f at the even positions  
 on the boudary except the corners 
 
 
.in
hk  sum of the values of f at odd positions  
I 4 
9  on the odd rows of the matrix except boundary rows  
ing
 sum of the values of f at even positions  
8   
 on the odd rows of the matrix except boundary rows  
 
eer
8 sum of the values of f at odd positions  
 on the even rows of the matrix except boundary rows  
 
gin
 sum of the values of f at even positions 

16  
 on the even rows of the matrix except boundary rows  
En
Examples on Numerical Integration

arn
2 dx

EG 7 : Compute the value of 1 x using
Le
(a) Trapezoidal rule (b) Simpson’s 1/3 rule

(c) Simpson’s 3/8 rule
w.
Solution :
Here h  0.25,
ww
1
y  f  x 
x
The tabulated values of  f  x  are
y
x0 x1 x2 x3 x4
x 1 1.25 1.50 1.75 2
1 1 1 1 1
f(x) 1  0.8  0.66667  0.57143  0.5
1 1.25 1.5 1.75 2
y0 y1 y2 y3 y4

(a) WKT, Trapezoidal rule is
b x0  nh h
a f  x  dx x0 f  x  dx  2  y0  y4   2  y1  y2  y3 
2 1  0.25
    dx  1  0.5   2  0.8  0.66667  0.57143 
1 x 2 
0.25 0.25
 1.5  4.0762  5.5762
2 2
 0.697025
(b) WKT, Simpson’s 1/3 rule is
b h
a f  x  dx  3  y0  yn   4  y1  y3   2  y2 
.in
2 1  0.25
    dx  1  0.5   4  0.8  0.57143  2  0.66667  
3 
ing
1 x
 0.693255
(c) WKT, Simpson’s 3/8 rule is
eer
b 3h
a f  x  dx  8  y0  y4   3  y1  y2   2  y3 
2 1  3  0.25 
gin
i.e.,    dx  1  0.5   3  0.8  0.66667   2  0.57143 

1 x 8
 0.66024
En
By actual integration,
21
1 x dx   loge x 1   ln 2  ln1  0.69315
2
arn
dx
10

Le
CW 8 : Evaluate 0 1  x by dividing the range into 8 equal parts by

w.

1
Here h  1.25, y  f  x  
ww
Solution : 1 x
(a) I = 2.51368 (b) I = 2.42200 (c) I = 2.41838
Actual integration, AI = 2.39790
1
HW 9 : Evaluate 0
x e x dx
taking 4 equal intervals by
Solution : (a) 1.02307 (b) 1.00017 (c) 0.87468, AI = 1

13 .

HW 10 : Calculate 0
3
sin x dx
taking 7 ordinates (6 intervals) using
a) Trapezoidal rule (b) Simpson’s 1/3 rule
Solution : (a) 1.33467 (b) 1.32612 (c)1.30516,AI=1.3333
2 dx
2
Eg 11 : Evaluate 0 x  4 using Romberg’s method. Hence obtain an
approximate value for π.
Solution :
To find I1
20 1
When h   1, y  f  x   2
.in
2 x 4
2 dx
Let I = 0 x 2  4
The tabulated values of y are
x 0 1 ing 2
eer
1 1
f  x  2 2
 0.25 0.2 0.125
x 4 0 4
gin
y0 y1 y2
Using Trapezoidal rule,
2 dx h
En
I1   2   y0  y2   2  y1  
0 x 4 2
1
arn
  0.25  0.125   2  0.2  

2
 0.3875
Le
To find I 2
20 1
h   0.5
w.
4 2
ww
x 0 0.5 1 1.5 2
1
f  x  2 0.25 0.23529 0.2 0.160 0.125
x 4
y0 y1 y2 y3 y4
2 dx h
I2   2   y0  y4   2  y1  y2  y3  
0 x 4 2
 0.25  0.25  0.125   2  0.23529  0.2  0.160  
 0.39136

To find I3
20 1
h  0.25
8 4
x 0 0.25 0.50 0.75 1 1.25 1.5 1.75 2
f  x 0.24 0.23 0.21 0.17 0.14

6 5 9 9 1 0.12
1 0.25 0.2 0.16
 1 2 1 1 5 5
x2  4 5 9 8 8 9
y0 y1 y2 y3 y4 y5 y6 y7 y8
.in
dx
2 h
I3     y0  y  
ing
x2  4 2
0
 0.25  0.125  
 0.125  
 2  0.24615  0.23529  0.21918  0.2  0.17918  0.16  0.14159  
eer
 0.39237
Romberg's formula for I1 & I 2 is
gin
 I2  I1   0.39136  0.39136 
I RM1,2  I 2   0.39136 
3 3
En
= 0.39265
Romberg's formula for I 2 & I3 is
 I3  I 2   0.39237  0.39237 
arn
I RM 2,3  I3   0.39237 
3 3
= 0.39271
Le
 I RM1,2  0.3927 & I RM 2,3  0.3927

Here I RM1,2 & I RM 2,3 are almost equal and I = 0.3927  1
w.
By actual integration,
ww
x2
2 dx 1 1
0 x 2  4  2  tan x 2  x 0  2  tan 1  tan  0 
1 1 1
1  
=   0     2
24  8

From 1 &  2  ,  0.3927
8
  = 8  0.3927   3.1416

15 .
1 1
CW 12: Using Romberg’s method, evaluate

0 1 x
dx
correct to 3
places
of decimals.
Solution :
I1  0.7083, I 2  0.6970, I3  0.6932
I RM1,2  0.6932, I RM 2,3  0.6931
 I = 0.693
Extras 13 : A curve passes through the points (1, 2), (1.5, 2.4), (2, 2.7),
(2.5, 2.8), (3, 3), (3.5, 2.6) & (4, 2.1). Obtain Area bounded by
.in
the curve, x axis between x = 1 and x = 4. Also find the volume
of solids of revolution by revolving this area about x – axis.
ing
Solution:
b 4
WKT, Area = a y dx  1 y dx
eer
4 h
Simpson's 1 rule, Area = 1 y dx   y0  y6   4  y1  y3  y5   2  y2  y4  
3 3
= 7.783
gin
b 4
Volume =   y 2 dx    y 2 dx
En
a 1
2
To find y :
x 1 1.5 2 2.5 3 3.5 4
arn
y 2 2.4 2.7 2.8 3 2.6 2.1

y2 4 5.76 7.29 7.84 9 6.76 4.41
Le
4 2 h 
 Volume =  y dx     y0  y6   4  y1  y3  y5   2  y2  y4   
1 3 
w.
= 64.07 cubic units.

Two points Guassian quadrature formula :
ww
 1  1   1 
 1 f  x  dx  f   
  3  3 
2 dx
Eg 14 : Evaluate
1 x
by using Guassian two point formula.
b  a  z  b  a 
Solution : Here a  1, b  2  1, so use x  ,
2
x   z  3 / 2  dx  dz / 2
 The above integral becomes

dx 21 dz 1 1 1
I =   dz   f  z  dz
1 x 1 z 3  1 z 3  1
 By Guassian two point formula

1  1   1 
1 f  z  dz  f  3
 f 
 3

 1   1 
  1  
 z  3 z  z  3  z  1
3 3
 0.693
.in
 2
0 log 1  x  dx
CW 15 : Using Guassian two point formula evaluate
Solution :
ing
1
I = log 1   1  z    dz
1  4  4
  1   1 
eer
   f   f  
4
  3  3 
 0.858
gin
2 dx
HW 16 : Evaluate 1 1  x
 3
by using Guassian two point formula.
En
Solution : I = 0.2544
Three points Guassian quadrature :

arn
 5  
 9  f  3 5  f

   8
3 5   f  0  
 9 
Le
2
1
 1  x3 dx
w.
Eg 17 : Evaluate 1 by using Guassian three point formula.

Solution :
b  a  z  b  a 
ww
Using the substitution x  , a  1, b  2

2
z 3
x
2
1 1 1
1  dz  1 1 8 1
I= 3  2  2 
 dz   dz
1  z 3   1  z 3
3 2 1 8   z  33
1   1  
 2   2 

17 .
1
4 4
  dz , f  z  
1 8   z  33 8+  z+3
3
5
  
   f  3 5  f 3 5   f  0
9  
8
9
 

5 8
  0.27505   0.11429 
9 9
 0.25439 ∵ 5  0.5555, 8  0.8888
 9 9 
I = 0.02544
11

.in
dt
CW 18 : Find t 0 by using Guassian three point formula.
ing
1 dx
Eg 19 : Evaluate
 0 1  x2
by two and three point Guassian quadrature
formula & hence find the value of  .
eer
Solution :
b  a  z  b  a  z 1 dz
x x  dx 
gin
2 2 2
1 dx 1 1 1
Now, I   
0 1  x2 1 2 2
 z 1
En
1  
 2 
1 1
arn
1
  f  z  dz , f  z   2
1 2  z 1
1  
 2 
Le
By Guassian two point formula

1  1   1 
w.
I= 1 f  z  dz  f 
  f 
3 
  0.7868
3
 1
ww
By Guassain three point formula

1 5
 8
  
I =  f  z  dz   f  3 5  f 3 5   f  0   0.7853   2 
1 9  9
1
1 dx
By actual integration, I = 0   tan 1  x  
1  x2  
0
 tan 1 1  tan 1  0    4   3

From  2  &  3 ,
0.7853   4
   3.147

 2
HW 20 : Find 0
sin x dx
by two & three point Guassian quadrature
formula.
Solution :
x    z  1 4
I = 0.9985  by two point formula 
I = 1.0000  by three point formula 
Double integrals using Trapezoidal rule & Simpson’s rule :

1.4 2.4 dx dy
.in
Eg 21 : Evaluate the integral
1 2 xy using Trapezoidal rule and
Simpson’s rule. Verify your results by actual integration.
ing
1
Solution : f  x, y   , x varies from 1, 1.4 
xy
eer
y varies from  2, 2.4 
Divide the range of x and y into 4 equal parts
gin
2.4  2 1.4  1
h  0.1, k  0.1
4 4
f  x, y  at the nodal points are given in the table :
En
The values of
x
2 2.1 2.2 2.3 2.4
y
arn
1 0.5 0.4762 0.4545 0.4348 0.4167

1.1 0.4545 0.4329 0.4132 0.3953 0.3788
1.2 0.4167 0.3698 0.3788 0.3623 0.3472
Le
1.3 0.3846 0.3663 0.3497 0.3344 0.3205

1.4 0.3571 0.3401 0.3247 0.3106 0.2976
w.
By Trapezoidal rule for double integration :

ww
Sum of values of f at the four corners 

 
hk   Sum of values of f at the nodes  
I +2  
4   on the boundary except the corners  
 
 4  Sum of the values at the interior nodes  

19 .
 
 
 0.5  0.4167  0.2976  0.3571  
 0.1 0.1   0.4762  0.4545  0.4348  0.3788  0.3472  0.3205  

 2  
4   0.3106  0.3247  0.3401  0.3846  0.4167  0.4545  
 
 4  0.4329  0.4132  0.3953  0.3623  0.3344  
  0.3497  0.3663  0.3698  0.3788 
 
 
 0.0614
By Extended Simpson’s rule :
hk
I   Sum of the values of f at the four corners  
.in
9  
 +2 Sum of the values of f at the odd positions on the boundary except the corners  
 
 +4 Sum of the values of f at the even positions on the boundary except the corners  
ing
  Sum of the values of f at the odd positions  
 +4  
  on the odd rows of the matrix except boundary rows  
 
  Sum of the values of f at the even positions  
 +8 on the odd rows of the matrix except boundary rows 
eer

   
  Sum of the values of f at the odd positions  
 8  
  on the even rows of the matrix except boundary rows  
gin
 
 16  Sum of the values of f at the even positions  
 
  on the even rows of the matrix except boundary rows  
 0.5  0.4167  0.2976  0.3571 
En
 
 2  0.4545  0.3472  0.3247  0.4167  
 
arn
 4  0.4762  0.4348  0.3788  0.3205  0.3106  

  0.3401  0.3846  0.4545 

 0.1 0.1  
   
Le
4 0.3788 
9  
 8  0.3968  0.3623 
w.
 8  0.3497  0.4132  
 
 16  0.3663  0.3344  0.4329  0.3953 
ww
 
 
0.01
  55.2116   0.0613
9
By actual integration :
1.4 2.4 1 1.4  2.4 1  1 1.4 2.4 1
1 2 xy dx dy  1  2 x dx  y dy  1  log x 2 y dy
  log 2.4  log 2  log y 1
1.4
 0.0613

 2  2
sin  x  y  dx dy
CW 22 : Evaluate 0 0 by using trapezoidal rule,
Simpson’s rule and also by actual integration.
Solution : Divide into 2 intervals (3 ordinates) on x & y

By Trapezoidal rule, I = 1.7975
Simpson’s rule, I = 2.0080
Actual integration, I = 2
0.5 0.5 sin xy
I=  0 1  xy dx dy by using Simpson’s rule with
0
HW 23 : Evaluate
step size 0.25.
.in
PART – A
ing
1. State the Romberg’s integration formula with h1 and h2 . Further, obtain
h
h1  h and h2  .
eer
the formula when 2
I1h22  I 2 h12
Solution : I  , where I1 the value of the integral with h1
h22  h12
gin
I 2 the value of the integral with h2

En
If h1  h & h2  h 2 we get
4 I 2  I1 1
I  I 2   I 2  I1 
arn
3 3
1
dx
 1  x2 .
Le
2. Use two-point Guassian quadrature formula to solve 1

Solution : Given interval is – 1 to 1 so we apply formula
w.
1
 1   1 
 f  x  dx  f  3   f  
3
ww
1
1
Here f  x  
1  x2
 1  1 1 3
f  1
 
 3  1  4  4
3  3 
 1  1 1 3
f  1
 
 3  1  4  4
3  3 

21 .
1
1 3 3 3
 2
dx     1.5
1 1  x
4 4 2
But actual integration
1
1 1
 1  x 2 dx   tan
1
x   tan 1 1  tan 1  1
 1
1
 tan 1 1  tan 1 1

 2 tan 1 1
 
 2
4 2
.in
 1.5708
Here the error due to two – point formula is 0.0708.
3.
ing
Write down the formulae for finding the first derivative using Newton’s
forward difference at x  x0 Newton’s backward difference at x  xn .
eer
Solution :
v v  v  1 2 v  v  1 v  2  3
y  x   yn  yn   yn   yn
gin
1! 2! 3!
v  v  1 v  2  v  3 4 x  xn
  yn , where v 
En
4! h
Here x  xn  v  0
y  x   yn
arn
y  x  h  y  x
y   x   yn  lim
h 0 h
Le
y   x   yn
w.
2
 x2
e dx
4. Evaluate 0 by two point Guassian quadrature formula.
ww
 x2
Solution : Here f (x) = e [Given range is not in exact form]
ba ba
Let x  z  Here a  0, b  2
2 2
20 20
 z
2 2
x  z 1 x  0  z  1
dx  dz x  2  z 1

2 1
x 2  z 12
 e dx  e dz
0 1
 1   1 
 f   f    1
 3  3
 z 1 2
Here f  z   e  
2
 1 
  1
 1 
f   e  3   e 0.1786  0.8364
 3
2
.in
 1 
  1
 1 
f   e  3   e 2.488  0.0831
 3
ing
2
 x2
 1  e dx  0.8364  0.0831
0
eer
 0.9195
5. Write down the Newton-Cote’s formula for the equidistant ordinates.

gin
Solution : The general Newton-Cote’s quadrature formula is

En
 n n  2 n  3  
2 n  n  2
a  nh 2
 f  x  dx  nh  1     3  ... f  a 
arn
a  2 12 24 
This is known as the general Gauss-Lagendre integration formula.
Putting n = 1 and omitting second and hyper differences in the above,
Le
we get
ah
h h
 f  x  dx  2  f  a   f  a  h   2  y1  y2 
w.
a
ww
which is the trapezoidal rule.
1
6. When do you apply Simpson’s 3 rule, and what is the order of the
1
error in Simpson’s 3 rule.
Solution : Let n = interval

23 .
1
Rule : Simpsons’ 3 rule → The number of ordinates is odd (or) the
intervals number is even.
2
The error in the trapezoidal formula is of the order h .
7. State Simpson’s one third rule.
Solution : Let DC be the curve y = f (x) and DA, CB be the terminal
ordinates. Let OA = a and OB = b. Divide AB into even number(say
2n) of equal parts, equal to h. Let x1 , x2 ,...x2 n 1 be the abscissae of the
points, A, A1 , B and y1 , y2 ,... y2 n 1 be the corresponding ordinates.
Then A 2n-1
.in
b
h
 y dx is approximately  3  A + 4B + 2C   1
a
ing
where A = y1  y2 n 1  sum of the first and last ordinates.
B = y2  y1  ... y2 n  sum of the even ordinates and
eer
C = y3  y5  ...  y2 n 1  sum of the remaining ordinates 1 is
known as Simpson's rule or Simpson's one-third rule.
gin
8. Write down two point Guassian quadrature formula to evaluate

1
 f  x  dx.
En
1
Solution : Two points Guassian quadrature formula is
1
arn

1   1 
 f  x  dx  f  f  
3  3
1
This formula is exact for polynomials upto degree 3.
Le
Unit III Assignment problems:

w.
Numerical Differentiation
1. Find f '(10) from the following data: (May 2011)
ww
x 3 5 11 27 34
f(x) –13 23 899 17315 35606
2. Find the first and second derivatives of y w.r.t x at x = 10 from the
data given below:
x 3 5 7 9 11
y 31 43 57 41 27
dy
3. Find dx at x = 1 from the following table:
x 1 2 3 4
y 1 8 27 64

4. Find the values of f '(4) and f ''(4) from the following data:
x 0 1 2 5
f(x) 2 3 12 147
5. Find cos 1.747  using the values given in the table below:
x 1.70 1.74 1.78 1.82 1.86
sin x 0.9916 0.9857 0.9781 0.9691 0.9584
6. Find sec(31  ) from the following data:
 31 32 33 34
tan  0.6008 0.6249 0.6494 0.6745
Numerical Integration
1
1
.in
 1  x2 dx
7. Evaluate 0 using Romberg’s method.
ing
2
1
 1  x3 dx
8. Evaluate 1 using Gaussian three point formula.
eer
1 1 1
0 0  x  y  1dx dy
9. Evaluate by using Trapezoidal rule taking
gin
h = 0.5 and k = 0.25.

2
x2  2 x  1
 1   x  14 dx
En
10. Evaluate 0 by Gaussian three point formula.

1 2 2 xy
0 1 dx dy
1  x 1  y 
2 2
arn
11. Evaluate by using Trapezoidal rule with

h = k = 0.25.
2.4 4.4
Le
Evaluate 2 4
xy dx dy
12. using Simpson’s rule given h = k = 0.1.
13. The velocity v of a particle at a distance S from a point on its path is
w.
given by the table below:

S(meter) 0 10 20 30 40 50 60
ww
v(m/sec) 47 58 64 65 61 52 38
Estimate the time taken to travel 60 meters by using Simpson’s three –
eighth rule.

14. By dividing the range into ten equal parts, evaluate by0 sin x dx
Trapezoidal and Simpson’s rule. Verify your answer with actual
integration.
10 dx
15.

Calculate 2 1  x by dividing the interval into eight equal parts and
11
log e .
hence find an approximate value of 3

UNIT IV
INITIAL VALUE PROBLEMS FOR ORDINARY
DIFFERENTIAL EQUATIONS
Single step methods: Taylor series method – Euler method for first order
equation – Fourth order Runge – Kutta method for
solving first and second order equations.
Multi step methods: Milne’s and Adam’s predictor and corrector methods.
Topic 1.1 : Taylor’s series method

dy
Given  y   f  x, y  with y  x0   y0
dx
.in
d2y f f dy
  
dx 2 x y dx
ing
ie., y   f x  f y  y  
llly find y , y ,...
eer
Taylor’s series expansion of y(x) above x  x0 is given by
 x  x0   x  x0 2  x  x0 3
y  x   y  x0   y   x0   y   x0   y   x0   ...
gin
1! 2! 3!
 x  x0   x  x0 2  x  x0 3
 y0  y0  y0  y0  ...
En
1! 2! 3!
 or 
arn
h h2 h3
y  x1   y1  y0  y0  y0  y0  ..., where x1  x0  h, h  x  x0
1! 2! 3!
h2 h3
Le
h
y  x2   y2  y1  y1  y1  y1 ..., where x2  x1  h, h  x  x0
1! 2! 3!
w.
Eg 1 : Using Taylor’s series method, find y at x = 0.1, if

dy
 x 2 y  1, y  0   1
ww
dx
2
Solution : Given y   x y  1, x0  0, y0  1, h  0.1
 Taylor’s series formula is
h h2 h3
y1  y0  y   y   y0  ...  1
1! 2! 3!
y  x2 y  1 y0  x0 2 y0  1  1
y   2 xy  x 2 y  y0  2 x0 y0  x02 y0  0
y   2  xy   y   x 2 y   y 2 x

2 Unit IV – Initial Value Problems For Ordinary Differential Equations
y0  2  x0 y0  y0   x 2 y0  y0 2 x0
2
y iv  2 y   2 xy   2 y   2 xy  iv
y0  2 y0  2 x0 y0  2 y0
 x y   y   2 x   2 y 
2
 2 xy0  x 2 y0  y   2 x0 
 2 y0
 6
0.1  0.1 2
 0.13  0.14
 1  y1  1   1   0   0   6 
1! 2! 3! 4!
 0.13  0.14
.in
ie., y  0.1  1   0.1    ...
3 4
ing
 0.900305
CW 2 : Using Taylor’s method, compute y (0.2) and y (0.4) correct to 4

eer
dy
  y   1  2 xy & y  0   0.
decimal places given dx
Solution : y (0.2) = 0.194752003, y (0.4) = 0.359883723
gin
HW 3 : Using Taylor’s series method, with the first five terms in the
expansion; find y (0.1) correct to 3 decimal places, given that
En
dy
 e x  y 2 , y  0   1.
dx
arn
Solution : y (0.1) = 1.0049891 ≃ 1.005 (correct to 3 decimal places)
Topic 1.2 : Taylor’s series method for simultaneous first order

Le
differential equations
dy dz
 f1  x, y, z  ,  f 2  x, y, z  with initial
w.
Given
dx dx
conditions y  x0   y0 , z  x0   z0
ww
EG 4 : Solve the system of equations

dy dz
 z  x2 ,  y  x with y  0   1, z  0   1
dx dx
by taking h  0.1 to get y  0.1 and z  0.1 .
Solution : Given x0  0, y0  1, z0  1
WKT, Taylor's series for y1 is y1  y  0.1
h h2 h3 h 4 iv
= y0  y0  y0  y0  y0  ...  1
1! 2! 3! 4!

& Taylor's series for z1 is z1  z  0.1

h h2 h3 h 4 iv
= z0  z0  z0  z0  z0  ...   2
1! 2! 3! 4!
dy y0  z0  x0 2  1  0 dz z0  x0  y0
y   z  x2 z   x y
dx 1 dx  0 1  1
y   z   2 x y0  1 z   1  y  z0  1  y0  2

y   z   2 y0  0 z   y  z0  1
y  z 
iv
y0iv  1 z iv  y  z0iv  0
.in
 0.1  0.12  0.13  0.14
 1  y  0.1  1  1  1   0  1
ing
1! 2! 3! 4!
= 1.1050  correct to 4 decimal places 
 0.1  0.12  0.13  0.14

eer
Now  2   z  0.1  1  1   2  1   0
1! 2! 3! 4!
= 1.110167
gin
≃ 1.1102  correct to 4 decimal places 

En
Topic 1.3 : Taylor’s series for II order differential equations

Eg 5 : By Taylor’s series, find y (0.1) and y (0.2) given that
y   y  xy , y  0   1, y   0   0.
arn
Solution : Given x0  0, y0  1, y0  0 & y   y  xy 

WKT, Taylor's series formula is
Le
x2 x3
y  x   y0  xy0  y0  y0  ...  1
2! 3!
w.
y   y  xy  y0  y0  x0 y0  1  0  0   1
ww
y   y   xy   y   2 y   xy  y0  0
y  2 y   xy   y   3 y   xy 
iv
y0iv  3
x2 x4
 1  y  x   1  0  1  0   3  ...
2! 4!
x2 x4
 1   ...
2 8

 0.12  0.14
y  0.1  1    ...
2 8
 1.0050125 ≃ 1.0050  correct to 4 decimal places 
 0.2 2  0.2 4
y  0.2   1    ...
2 8
 1.0202
Topic 2.1 : Euler method

Given y   f  x, y  , x0 , y0 , h
.in
Euler algorithm is
ing
y1  y0  hf  x0 , y0 
y 2  y1  hf  x1 , y1 
y3  y2  hf  x2 , y2 
eer
⋮
y n  yn 1  hf  xn 1 , yn 1 
gin
Eg 6 : Using Euler’s method, find y (0.2), y (0.4) and y (0.6) from

En
dy
 x  y, y  0   1 with h  0.2.
dx
arn
Solution : Given y  0   or  y0   1  x0  0

& x1  0.2, x2  0.4, x3  0.4, h  0.2
Le
y  x  y
We have to find y  0.2   y1
w.
Now, by Euler algorithm,

y1  y0  hf  x0 , y0 
ww
ie., y  0.2   1   0.2  x  y    x  x0   1  0.2  x0  y0   1   0.2  0  1

 y y 
 0
 y  0.2   1.2  y1
y2  y1  hf  x1 , y1   1.2   0.2   x1  y1 
 1.2  0.2  0.2  1.2  1.48  y  0.4   1.48
y3  y2  hf  x2 , y2   1.48   0.2   x2  y2 
 1.856  y  0.6   1.856

 The result of the problem:

x y
x0  0 y0  1
x1  0.2 y1  y0  hf  x0 , y0   1.2
x2  0.4 y2  y1  hf  x1 , y1   1.48
x3  0.4 y3  y2  hf  x2 , y2   1.856
dy
 y  ex , y  0  0
CW 7 : Solve dx for x = 0.2, 0.4 by using Euler’s
method.
.in
Solution : y1  0.2, y2  0.484281
ing
HW 8 : Using Euler’s method, find the solution of the initial value problem
dy
 log  x  y  , y  0   2
eer
dx at x = 0.2 by assuming h = 0.2.
Solution : y (0.2) = 2.0602
gin
Topic 2.2 : Modified Euler method

 h 1 
yn 1  yn  hf  xn  , yn  hf  xn , yn  
 2 2 
En
 h h 
n  0, y1  y0  hf  x0  , y0  f  x0 , y0  
 2 2 
arn
 h h 
n  1, y2  y1  hf  x1  , y1  f  x, y  
 2 2 
Le
 h h 
y3  y2  hf  x2  , y2  f  x2 , y2  
 2 2 
w.
Eg 9 : Solve y   1  y, y  0   0 by modified Euler method.

Solution : Given y   f  x, y   1  y, x0  0, y0  0
ww
Let h  0.1,  x1  0.1, x2  0.2, x3  0.3

We have to find y1 , y2 , y3 .
By Modified Euler’s method:
 h 1 
yn 1  yn  hf  xn  , yn  hf  xn , yn  
 2 2 
 h h 
y1  y0  hf  x0  , y0  f  x0 , y0  
 2 2 
 Here f  x0 , y0   f  x  x0 , y  y0   1  y0  1  0  1

 0.1 0.1 
Therefore y1  0   0.1 f 0  , 0 1
 2 2 
 0.1 f  0.05,0.05   0.1 1  y 
 xy0.05
0.05 
 1  0.05
y1  0.095
 h h 
Now y2  y1  hf  x1  , y1  f  x1 , y1  
 2 2 
 Here f  x1 , y1   1  y1  0.905
 0.1 0.1 
.in
i.e., y2  0.905   0.1 f 0.1  , 0.095  f  0.1, 0.095  
 2 2 
y2  0.18098
 h h
Now y3  y2  hf  x2  , y2  f  x2 , y2  
 2 2

 ing
eer
 Here f  x2 , y2   1  y2  1  0.18098  0.81902 
 0.1 0.1 
gin
i.e., y3  0.18098   0.1 f 0.2  , 0.18098  f  0.2, 0.18098  

 2 2 
y3  0.258787
En
 y0  0, y1  0.095, y2  0.18098, y3  0.258787

arn
CW 10 : By using modified Euler method, find

dy x  y
y 1.8  if  , y  2  1  Hint h  0.2
dx x  y
Le
Solution : y (1.8) = 0.9349

w.
dy
HW 11 : Given  y  x 2  0, y  0   1, y  0.1  0.9052,
dx
ww
y  0.2   0.8213, find correct to four decimal places y  0.3 ,

using modified Euler's method.
 Hint y   x 2  y 
Solution :  
 h h 
y  0.3  y3  y2  hf  x2  , y2  f  x2 , y2    0.7493
 2 2 

Topic 3.1 : Fourth order Runge – Kutta method for solving I order
Differential Equations[Single step method]
dy
Given  y   f  x, y 
dx
& y  x0   c0 ,
Now y  x1   ?  c1 , y  x2   ?  c2 , y  x3   ?  c3 ,...
General
h(  Interval length)
xn  known xn 1  known
.in
yn  known yn 1  yn  yn 1  to find 
ing
To find f  xn 1   y  xn 1   yn 1 :
Fourth order Runge – Kutta Algorithm :
yn 1  y  xn  h   y  xn   yn 1  yn  yn 1
eer
1
where yn 1   k1  2k2  2k3  k4 
6
gin
where k1  hf  xn , yn 
 h k 
k2  hf  xn  , yn  1 
En
 2 2
 h k 
arn
k3  hf  xn  , yn  2 
 2 2 
k4  hf  xn  h, yn  k3 
Le
To find f  x1   y  x1   y1 :
k1  hf  x0 , y0 
w.
 h k 
k2  hf  x0  , y0  1 
ww
 2 2
 h k 
k3  hf  x0  , y0  2 
 2 2 
k4  hf  x0  h, y0  k3 
1
y1   k1  2k2  2k3  k4 
6
 y1  y  x0  h   y  x0   y1  y0  y1

dy
 x  y 2 , y  0   1,
Eg 12 : Given dx find y (0.1) & y (0.2) by Runge –
Kutta method for IV order.
Solution : Given y   f  x, y   x  y , h  0.1
2
& x0  0, y  x0  0   1.
To find y (0.1) :
k1  hf  x0 , y0    0.1 f  0.1   0.1 0  12   0.1
 
 h k 
k2  hf  x0  , y0  1    0.1 f 0  0.05, 1  0.05  
.in
 2 2
  0.1 0.05  1.05    0.11525
2
ing
 
 h k   0.11525 
k3  hf  x0  , y0  2    0.1 f 0.05,1 
 2 2  2 
eer
  0.1 0.05  1.057625    0.116857
2
 
gin
k4  hf  x0  h, y0  k3    0.1 f  0.05,1  0.116857 

2
  0.1 0.05  1.116857    0.134737
 
En
1 1
y   k1  2k2  2k3  k4    0.1  2  0.1  2  0.11525    0.116857  
6 6
arn
 0.11649
 y  0.1  y1  y0  y  1  0.11649  1.11649  y1  1.11649
Le
To find y  0.2   y2 : (Second interval)

k1  hf  x1 , y1    0.1 f  0.1,1.11649   0.1 0.1  1.116492  
w.
 0.1347
ww
 h k   0.1347 
k2  hf  x1  , y1  1    0.1 f 0.1  0.05,1.11649 
 2 2  2 
  0.1 f  0.15,1.18385  0.1552
 h k   0.1552 
k3  hf  x1  , y1  2    0.1 f 0.1  0.05,1.11649 
 2 2   2 
  0.1 f  0.15,1.1941  0.1576
k4  hf  x1  h, y1  k3    0.1 f  0.1  0.1,1.11649  0.1576
  0.1 f  0.2,1.2741  0.1823

1
y   k1  2k2  2k3  k4   0.1571
6
y2  y1  y  1.11649  0.1571  1.27359
i.e.,
x 0 0.1 0.2
y 1 1.11649 1.27359
CW 13 : Using R. K. method of fourth order solve for

dy 2 xy  e x
for x  1.4, from  with y  0 at x  1.
dx x 2  xe x
.in
Solution : y1  0.140  Here y2  y 1.4   0.27  , h  0.2
HW 14 : Solve y for x = 0.2, 0.4 from the given O.D.E. ing

dy
dx
 x2  y
and
eer
y = 0.8 at x = 0.
Solution : y1  y  0.2   0.99029, y2  y  0.4   0.120828
gin
Topic 3.2 : Fourth Order Runge – Kutta method for solving II order
differential equation
En
Eg 15 : Consider the II order initial value problem

arn
y   2 y   2 y  e 2t sin t with y  0    0.4 and y   0    0.6 using fourth

order Runge – Kutta method, find y (0.2), z (0.2).
Solution : Let t  x
Le
 y   2 y   2 y  e 2 x sin x, y  x  0   0.4
y   x  0   0.6
w.
 y   2 y   2 y  e 2 x sin x  1
ww
Here h  0.2
Setting y   z , y   z 
 1  z   2 z  2 y  e 2 x sin x
dy
Let f1  x, y, z    z,
dx
d2y dz
& f 2  x, y , z   2
  2 z  2 x  e 2 x sin x
dx dx

k1  hf1  x0 , y0 , z0  ℓ 1  hf 2  x0 , y0 , z0 
  0.2   z0    0.2   2 z0  2 y0  e 2 x0 sin x0 
 
  0.2  0.6   0.12 2 0 

  0.2   2  0.6   2  0.4   e sin  0  
 
 0.136
 h k ℓ   h k ℓ 
k2  hf1  x0  , y0  1 , z0  1  ℓ 2  hf 2  x0  , y0  1 , z0  1 
 2 2 2  2 2 2
 0.2   0.2 0.2 
0  2 ,  0  2 , 0.4  2 , 
    0.2  f 2  
  0.2   0.4  
0.12 
,  0.6  0.136 
.in
 2   2 
 
 0.6   0.136    0.2  f 2  0.1,  0.46,  0.532
 2 
ing
 2  0.532   2  0.46  
 0.1064   0.2   2 0.1 
  e   sin  0.1 
eer
  0.2   1.064  0.92  0.1294
 0.00292
ℓ   ℓ 
gin
 h k h k
k3  hf1  x0  , y0  2 , z0  2  ℓ 3  hf 2  x0  , y0  2 , z0  2 
 2 2 2  2 2 2
 0.2   0.2 0.1064 
0  2 , 0  2 , 0.4  2 , 
En

    0.2  f 2  
 0.1064   0.00292 
  0.2  f1 0.4  ,  0.6 
   2 
arn
2
 
 0.6  0.00292    0.2  f 2  0.1, 0.4532, 0.60146
 2   2  0.60146   2  0.4532  
Le
  0.2  f1  0.1, 0.4532, 0.60142   0.2   2 0.1 

  e   sin  0.1 
  0.2   0.60146  0.1203
  0.2   1.20292  0.9064  0.12194
w.
 0.0105
k4  hf1  x0  h, y0  k3 , z0  ℓ 3  ℓ 4  hf 2  x0  h, y0  k3 , z0  ℓ 3 
ww
0  0.2,  0  0.2, 0.4  0.1203, 

  0.2  f 2 
  0.2  f1  0.4  0.1203,    0.6  0.0105 
 0.6  0.0105   2  0.6105   2  0.5203 
  0.2  f1  0.2, 0.5203, 0.6105   0.2   2 0.2 
  e   sin  0.1 
  0.2   0.6105
  0.2   1.221  1.0406  0.29638
 0.1221
 0.0825

11 .
1 1
y   k1  2k1  2k2  2k3  z   ℓ 1  2ℓ 2  2ℓ 3  ℓ 4 
6 6
 0.1159  0.03194
y1  y  0.2   y0  y z1  z  0.2   z0  z
 0.5159  0.2806
CW 16 : Given y   xy   y  0, y  0   1, y   0   0, find the value of y(0.1)

by using R. K. method of fourth order.
Solution : y (0.1) = 0.9950
HW 17 : Given y   x  y    y  0, y  0   1, y   0   0, find the value of

22
.in
y(0.2) by using R. K. method of fourth order.
ing
Solution : y (0.2) = 0.9801
Topic 3.3 : R. K. Method for simultaneous first order differential

eer
equations
dy dz
 f1  x, y, z  &  f 2  x, y , z 
Solving the equation dx dx with
gin
the initial conditions y  x0   y0 , z  x0   z 0 .
Now starting from  x0 , y0 , z0  with increments,

y & z
in y & z
En
respectively. Use formula

k1  hf1  x0 , y0 , z0  ℓ 1  hf 2  x0 , y0 , z0 
arn
 h k ℓ   h k ℓ 
k2  hf1  x0  , y0  1 , z0  1  ℓ 2  hf 2  x0  , y0  1 , z0  1 
 2 2 2  2 2 2
Le
 h k ℓ   h k ℓ 
k3  hf1  x0  , y0  2 , z0  2  ℓ 3  hf 2  x0  , y0  2 , z0  2 
 2 2 2   2 2 2 
w.
k4  hf1  x0  h, y0  k3 , z0  ℓ 3  ℓ 4  hf 2  x0  h, y0  k3 , z0  ℓ 3 
1 1
ww
y   k1  2k2  2k3  k4  z   ℓ 1  2ℓ 2  2ℓ 3  ℓ 4 
6 6
y1  y0  y z1  z0  z
dy dz
 xz  1,   xy
Eg 18 : Solving the system of differential equation dx dx
for x = 0.3 using fourth order R. K. method, the initial values are
x = 0, y = 0, z = 1.
Solution : Given x0  0, y0  0, z0  1, h  0.3.

f1  x, y, z   xz  1 f 2  x, y, z    xy
k1  hf1  x0 , y0 , z0    0.3  x0 z0  1 ℓ 1  hf 2  x0 , y0 , z0    0.3  0  0  
  0.3 0  1  0.3 0
 h k ℓ   h k ℓ 
k2  hf1  x0  , y0  1 , z0  1  ℓ 2  hf 2  x0  , y0  1 , z0  1 
 2 2 2  2 2 2
 0.345  0.007
 h k ℓ   h k ℓ 
k3  hf1  x0  , y0  2 , z0  2  ℓ 3  hf 2  x0  , y0  2 , z0  2 
 2 2 2   2 2 2 
.in
 0.3448  0.0078
k4  hf1  x0  h, y0  k3 , z0  ℓ 3  ℓ 4  hf 2  x0  h, y0  k3 , z0  ℓ 3 
ing
 0.3893  0.031032
1 1
y   k1  2k2  2k3  k4   0.34482 z   ℓ 1  2ℓ 2  2ℓ 3  ℓ 4   0.01011
eer
6 6
y1  y0  y  0  0.34482 z1  z0  z  1   0.01011
gin
 y  0.3  0.34482  z  0.3  0.98989

CW 19 : Solve the simultaneous differential equation
dy dz
En
 2 y  z,  y  3 z , y  0   0, z  0   0.5
dx dx
for y (0.1) and z (0.1) using R. K. method of fourth method.
arn
Solution : Given x0  0, y0  0, z0  0.5, h  0.1

 y  0.1  0.04814, z  0.1  0.37263.
Le
HW 20 : Using the R. K. method, tabulate the solution of the system

dy dz
 x  z,  x  y, y  0, z  0,
w.
dx dx
when x = 0 at intervals of h = 0.1 from x = 0.0 to x = 0.2.
ww
Solution : Given x0  0, y0  0, z0  1, h  0.1

 y  0.1  0.1050, z  0.1  0.9998
y  0.2   0.2199, z  0.2   0.9986
Topic 4 : Milne’s predictor and corrector methods (multistep method)
Milne’s predictor formula
4h
yn 1, p  yn 3   2 yn  2  yn 1  2 yn 
3
Milne’s corrector formula
h
yn 1, c  yn 1   yn 1  4 yn  yn 1 
3

13 .
dy 1
  x  y  , y  0   2, y  0.5   2.636,
Eg 21 : Given dx 2
y 1  3.595, y 1.5   4.968 by Milne’s method to find y (2).
Solution : Given
x0  0 x1  0.5 x2  1 x3  1.5 x4  2
y0  2 y1  2.636 y2  3.595 y3  4.968 y4  ?
Here h = 0.5
1
y    x  y   f  x, y 
Given 2
.in
By Milne's predictor formula,
4h
 2yn 2  yn 1  2 yn 
ing
yn 1, p = yn 3 +
3
4h
n  3, y31, p = y0 + 2y1  y2  2 y3   1
eer
3
1 1
Now, y1 =  x1  y1  = 0.5+2.636 =1.568
2 2
gin
1 1
y2 =  x2  y2  = 1+3.595 =2.2975
2 2
En
1 1
y3 =  x3  y3  = 1.5+4.968 =3.234
2 2
4  0.5 
arn
1  y4, p = 2+  2 1.568   2.2975  2  3.234  

3 
= 6. 871  = y4 , say 
Le
By Milne’ corrector formula,

h
yn 1, c  yn 1   yn 1  4 yn  yn 1 
w.
3
h
n  3, y4, c  y2   y2  4 y3  y4    2
ww
3
1 1
Now y4   x4  y4    2  6.871  4.4355
2 2
0.5
 2   y4, c  3.595   2.2975  4  3.234   4.4355
3
0.5
 3.595  19.669   6.8732
3
 Corrected value of y at 2 is y  2  = 6.8732.

dy 3
 x  y, y  0   2, y  0.2   2.073,
CW 22 : Given dx the values of
y  0.4   2.452, y  0.6   3.023 are got by R.K. Method. Find
y  0.8  by Milne’s method.
Solution : y4, p  4.1659, y4, c  3.7953

HW 23:Solve y   x  y , 0  x  1, y  0   0, y  0.2   0.02,
2
y  0.4   0.0795, y  0.6   0.1762 by Milne’s method to find y (0.8) and y (1).
Solution : y (0.8) = 0.3046, y (1) = 0.4515.
.in
Topic 5 : Adam’s predictor and corrector methods
(efficient multistep method) [Adam’s-Bashforth method]
ing
Adam’s predictor formula :
h
yn 1, p  yn  55 yn  59 yn 1  37 yn  2  9 yn  3 
24
eer
Adam’s corrector formula :
h
yn 1, c  yn  9 yn 1  19 yn  5 yn 1  yn  2 
24
gin
 x 2 1  y  , y 1  1, y 1.1  1.233, y 1.2   1.548,

dy
Eg 24 : Given dx
En
y 1.3  1.979, evaluate y (1.4) by Adam’s-Basforth method.

Solution : Given
arn
x0  0 x1  1.1 x2  1.2 x3  1.3 x4  1.4

y0  1 y1  1.233 y2  1.548 y3  1.979 y4  ?
Here h = 0.1
Le
By Adam’s predictor formula,

h
yn 1, p  yn  55 yn  59 yn 1  37 yn  2  9 yn  3 
w.
24
h
n  3, y4, p  y3  55 y3  59 y2  37 y1  9 y0   1
ww
24
Here y0  x0 2 1  y0   12 1  1  2
y1  x12 1  y1   1.1 1  1.233  2.70193
2
y2  x2 1  y2   1.2  1  1.548  3.60912

2
y3  x3 1  y3   1.3 1  1.979  5.0345

2
0.1
 1  y4, p 1.979  55  5.0345   59  3.60912   37  2.70193  9  2  
4 
 2.5870451

15 .
By Adam’s corrector method,

h
yn 1, c  yn  9 yn 1  19 yn  5 yn 1  yn  2 
24
h
 n  3 , y4, c  y3  9 y4  19 y3  5 y2  y1    2
24
y4   x4  1  y4   1.4  1  2.5871  7.030716
2 2
0.1
  2   y4, c 1.979  9  7.030716   19  5.0345   5  3.60912    2.70193 
24 
 2.5772844.
.in
2
CW 25 : Using Adam’s Bashforth method, find y (4.4) given 5 xy   y  2,
y (4) =1, y (4.1) =1.0049, y (4.2) =1.0097 & y (4.3) =1.0143.
ing
2  y2
y  ,  y4, p  1.0186 & y4, c  1.0187.
Solution : Here 2x
CW 26 : Evaluate y (0.9), using Adam Bashforth’s predictor-corrector
eer
1
dy
 xy 3 ,
method, given that dx y (1)=1, y (1.1)=1.10681, y (1.2)=1.22787
gin
and y (1.3)=1.36412.
Solution : Here h = – 0.1,  y4, p  0.906518 & y4, c  0.906520.
En
arn
Le
w.
ww

PART – A
dy
 x  y,
1. Use Euler’s method to find y (0.2) and y (0.4) given dx
y (0) = 1 with h = 0.2.
Solution : f  x, y   x  y, x0  0, y0  1, x1  0.2, x2  0.4
By Euler algorithm,
y1  y0  hf  x0 , y0   1   0.2   x0  y0   1   0.2   0  1
i.e., y1  y  0.2   1.2
y2  y1  hf  x1 , y1   1.2   0.2   x1  y1   1.2   0.2   0.2  1.2
.in
 1.2  0.28
i.e., y2  y  0.4   1.48
ing
y3  y2  hf  x2 , y2   1.48   0.2   x2  y2   1.48   0.2   0.4  1.48
 1.48  0.376
eer
i.e., y3  y  0.6   1.856
2. Write the Adam-Bashforth predictor and corrector formulae.
gin
Solution : Adam’s predictor and corrector formulas are

h
yk 1, p  yk  55 yk  59 yk 1  37 yk  2  9 yk  3 
En
24
h
yk 1, c  yk  9 yk 1  19 yk  5 yk 1  yk  2 
24
arn
dy
 x  y,
Le
3. Find y (0.1) by using Euler’s method given that dx

y (0) = 1.
Solution : Given,
w.
f  x, y   x  y, x0  0, y0  1, x1  0.1, h  0.1
By Euler algorithm,
ww
y1  y0  hf  x0 , y0 
 1   0.1  x0 +y0 
 1   0.1  0  1
 1  0.1
 1.01
i.e., y  0.1  1.01

17 .
4. What are multistep methods? How are they better than single step
methods?
Solution : 1. Milne’s predictor – correction method.
2. Adams-Bashforth predictor – correction method.
In the single step methods, it is not possible to get any information
about truncation error.
In the multi step methods, it is possible to get easily a good estimate of
the truncation error.
5. Using Taylor series method, find y (1.1) given that y   x  y, y 1  0.
Solution : Given y   x  y, x0  1, y0  0, h  0.1
y  x  y y0  x0  y0  1  0  1
.in
y   1  y  y0  1  y0  1  1  2
y   y  y0  y0  2
y iv  y 
h h2 h3
y0iv  y0  2
h 4 iv ing
eer
y1  y0  y0  y0  y0  y0  ...
1! 2! 3! 4!
 0.1  0.1  0.1
2 3 4
0.1
 y 1.1  0  1   2   2   2   ...
gin
1! 2! 3! 4!
 0.1  0.1
3 4
En
 0.1   0.1 
2

3 1.2
 0.1  0.01  0.0003  0.0000083
arn
 0.1103083
 0.1103  correct to four decimals 
Le
Find y (0.2) for the equation y   y  e , given that y (0) = 0, by using

x
6.
Euler’s method.
w.
Solution : Given, f  x, y   y  e , x0  0, y0  0, h  0.2

x
ww
By Euler algorithm,
y1  y0  hf  x0 , y0 
 0  0.2 f  0, 0 
 0.2 0  e0   0.2
i.e., y  0.2   0.2

dy
 f  x, y  with y  x0   y0 .
7. State Euler method to solve dx
Solution : yn 1  yn  hf  xn , yn  when n  0, 1, 2, ...
This is Euler algorithm. It can also be written as
y  x  h   y  x   hf  x, y 
.in
ing
eer
gin
En
arn
Le
w.
ww

19 .
Unit IV Assignment problems

dy
 x 2 y  1, y (0)  1
1. Using Taylor series method find y at x = 0.1 if dx
2. Using modified Euler’s method find y at x = 0.1 if

dy
 x 2  y 2 , y (0)  1
dx
3. Using Runge – Kutta method of fourth order, solve

dy y 2  x 2
 with y (0)  1
dx y 2  x 2
.in
at x = 0.2
Given y   xy   y  0, y (0)  1, y (0)  0 ,find the value of y(0.1) by
ing
4.
using R.K.method of fourth order.
eer
5. Using R. K method of order 4,find y for x =0.1, 0.2, 0.3 given that
dy
 xy  y 2 , y (0)  1
dx also find the solution at x = 0.4 using Milne’s
gin
method.
2
Using Milne’s method to find y(4.4) given that 5 xy   y  2,
En
6.
y (4)  1, y (4.1)  1.0049, y (4.2)  1.0097, y (4.3)  1.0143 .
arn
2
7. Given y   xy  y , y (0)  1 ,find y(0.1) by Taylor’s method, y(0.2) by
Euler’s method, y(0.3) by Runge – Kutta method and y(0.4) by Milne’s
Le
method.
w.
dy
 y  x 2  1, y (0)  0.5
8. Consider dx
ww
(i) Using the modified Euler’s method, find y(0.2)

(ii) Using R.K. method of order 4, find y(0.4) and y(0.6)
(iii) Using Adam’s – Bashforth predictor corrector method, find y(0.8).

UNIT V BOUNDARY VALUE PROBLEMS IN ORDINARY AND

PARTIAL DIFFERENTIAL EQUATIONS
Finite difference solution of second order ordinary differential equation –
Finite difference solution of one dimensional heat equation by explicit and
implicit methods – One dimensional wave equation and two dimensional
Laplace and Poisson equations.
Topic 1: Finite difference solution of second order ordinary
.in
differential equation
Consider the problem
y ''  x   f  x  g '  x   g  x  y  x   r  x 
ing
with the boundary conditions y  x0   a and y  xn   b.
eer
Foe solving the equation, replace y '' and y ' by the formulae
y  2 yi  yi 1
y ''  i 1
gin
h2
y  yi 1
y '  i 1
En
2h
d2y
yx
arn
Eg (1) : Solve the differential equation dx 2 with

1
y  0   0, y 1  0 with h 
4 by finite difference method.
Le
Solution : Given differential equation

d2y
yx
w.
dx 2
d 2 y  x
ww
i.e.,  y  x  x  1
dx 2
Using the central difference approximation, we have
y  2 yi  yi 1
y ''  i 1   2
h2
Substitute (2) in (1), we get
yi 1  2 yi  yi 1
 yi  xi
h2
yi 1  2 yi  yi 1
 yi  xi
h2

2 Unit V – B.V.P. in ordinary and Partial differential equations
yi 1  2 yi  yi 1  h 2 yi  h 2 xi
1 1  1
yi 1  2 yi  yi 1  yi  xi ∵ h  
16 16  4
16 yi 1  33 yi  16 yi 1  xi   3
Put k  1, 2,3 in  3 , we get

1
16 y0  33 y1  16 y2  
4

1
16 y1  33 y2  16 y3     4
.in
2
3
16 y2  33 y3  16 y4  
4
Given y 0  0 i.e., y  x0  0   0 i.e., y0  0 
 
ing

y 1  0 i.e., y  x4  1  0 i.e., y4  0 
  5
eer
1 1 1 3
Since h  , we have x1  x0  h  , x2  x1  h  , x3  x2  h 
4 4 2 4
gin
Substitute (5) in (4), we get

1
0  33 y1  16 y2    6
4
En
1
16 y1  33 y2  16 y3   7
2
arn
3
16 y2  33 y3  0   8
4
Le
Solving (6) & (8), we get

1
y1  y3   9
66
w.

131
ww
256 y1  833 y3   10 

4
y3  0.05004
Substitute y3 in (9), we get
y1  0.03488
Substitute y1 & y3 in (6), we get
y2  0.05632
 Solution of the given differential equation is
y1  0.03488, y2  0.05632, y3  0.05004

CW (2) : Solve by finite difference method, the boundary value problem

y ''  x   y  x   2, 1
y  0   0, y 1  1, taking h  .
when 4
1 2 3
y1  y    0.0451, y2  y    0.2183, y3  y    0.5301
Solution : 4 4 4
CW (3) : Solve by finite difference method, the boundary value problem
y ''  x   3 y '  x   2 y  x   0 when y  0   2, y 1  10.1.
Solution : y  0.5   0.4367
HW (4) : Using the finite difference method, find y(0.25), y(0.5) and
y(0.75) satisfying the differential equation y ''  x   y  x   x
.in
subject to the boundary conditions y  0   0, y 1  2.
ing
1 2 3
y1  y    0.5443, y2  y    1.0701, y3  y    1.5604
Solution : 4 4 4
eer
Classification of PDE of second order:
 2u  2u  2u  u u 
A  x, y  2  B  x, y   C  x, y  2  f  x, y , ,   0
gin
x x y y  x y 
If B 2  4 AC  0  Elliptic form 
En
B 2  4 AC  0  Parabola form 
B 2  4 AC  0  Hyperbola form 
arn
Eg: (1): Classify f xx  2 f xy  0, x  0, y  0.

Solution: Here A  1, B  2, C  0.
Le
B 2  4 AC  4  0
 Given equation is Hyperbic form
w.
Eg: (2): Classify xf xx  yf xy  0, x  0, y  0.

Solution: Here A  x, B  0, C  y.
ww
B 2  4 AC  4 xy  0 when x  0, y  0
 Given equation is Elliptic form
Eg: (3): Classify u xx  2u xy  u yy  0.
Solution: Here A  1, B  2, C  1.
B 2  4 AC  0
 Given equation is Parabolic form
2
 2
Eg: (4): Classify x f xx  1  y f yy  0, 1  y  1,   x  .

Solution: Here A  x , B  0, C  1  y  .
2 2
B 2  4 AC  4 x 2 1  y 2   4 x 2  y 2  1 .
  x  
x 2 is always ‘+’ in
2
In 1  y  1, y  1 is ‘–’.
B 2  4 AC  ve  x  0 
 Given equation is Elliptic form
If x  0, B  4 AC  0  Given equation is Parabolic form
2
If y  1, B  4 AC  0  Given equation is Hyperbic form

2
.in
Eg: (5):
Hyperbolic
ing
Parabolic Type Elliptic Type
Type
 2u  2u  2u  2u
2 2  0
eer
u  u t 2
x 2
x 2
y 2
2 2
t x (One (Laplace equation in
(One dimensional Heat equation) dimensional Two dimensional
gin
Wave equation) Heat equation)

↙ ↘  2u  2u
  f  x, y 
En
Explicit form Implicit form 2

x y 2
 Bender  Schmidt's   Crank  Nicolson's  (Poisson equationin
   
arn
 method   method  Two dimensional

Heat equation)
Le
Topic 2: Finite difference solution of one dimensional heat equation by

explicit method(Bender–Schmidt’s method)
w.
u  2u
2 2
One dimensional heat equation is t x
ww
1
2 
Put a
u 1  2u

t a x 2
u  2u
i.e., a 
t x 2

(i) Bender–Schmidt’s difference equation is

ui , j 1    ui 1, j  ui 1, j   1  2  ui , j  1
k  1
 2
Here   
where ah  2
1
0 
Equation (1) is called explicit formula. This is valid if 2
(i) Bender–Schmidt’s difference equation is
1
ui , j 1   ui 1, j  ui 1, j    2
2
1 k a
   2 i.e., k  h 2
.in
where 2 ah 2
Equation (2) is called Bender–Schmidt’s recurrence equation.
Eg (5) : Solve x
 2u
2
u
2
t ing
 0 given u  0, t   0, u  4, t   0,
eer
u  x, 0   x  4  x  . Assume h = 1. Find the values of u upto t = 5.
 2u u  2u u
gin
2 is of the form 2  a a2

Solution : Given x
2 t x t
a 2 2
k  h 2  k  1  1
En
By Bender–Schmidt’s relation, 2 2
Step size in time(t) = k = 1 = t
Step size in x = h = 1 = x
arn
ui , j 1
ui , j 1   ui 1, j  ui 1, j 
The values of are calculated by 2 as follows:
Le
 x direction 
h
0 1 2 3 4
t
w.
u 1, 0  u  2, 0  u  3, 0 
u (0, t ) u  4, t 
ww
0 
 1 4 1  
 2 42   3  4  3
0 0
 3 4 3
t 04 33 40
1 0 2 3 2 0
dir. 2 2 2
 2 0 1.5 2 1.5 0
3 0 1 1.5 1 0
4 0 0.75 1 7.5 0
5 0 0.5 0.75 0.5 0

 2 f f
  0, f  0, t   f  5, t   0, f  x,0   x 2  25  x 2  ,
2 t
Eg (6) : Given x
find f in the range takingn h = 1 and upto 5 seconds.
 2 f f 2 f f
 is of the form a  a  1, h  1
Solution : Given  x 2  t x 2 t
a 1 2 1
k  h 2  k  1  k 
By Bender–Schmidt’s relation, 2 2 2
1  t
Step size in time(t) = k = 2 =
Step size in x = h = 1 = x
.in
Given f  0, 0   0, f 1, 0   12  25  12   24, f  2, 0   22  25  22   84,
f  3, 0   32  25  32   144, f  4, 0   42  25  42   144, f  5, 0   0
The values of
ui , j
are calculated by
1
2 ing
ui , j 1   ui 1, j  ui 1, j 
as follows:
eer
 x direction 
i
0 1 2 3 4 5
t=jk
gin
 0 0 24 84 144 144 0
t 0.5 0 42 84 114 72 0
1 0 42 78 78 57 0
En
d 1.5 0 39 60 67.5 39 0
i 2 0 30 53.25 49.5 33.75 0
arn
r
e 2.5 0 26.625 39.75 43.5 24.75 0
c 3 0 19.875 35.0625 32.25 21.75 0
t 3.5 0 17.5312 26.0625 28.4062 16.125 0
Le
i
o 4 0 13.0312 22.9687 21.0938 14.2031 0
n 4.5 0 11.4843 17.0625 18.5859 10.5469 0
w.
 5 0 8.5312 15.0351 13.8047 9.2929 0

CW (7) : Solve ut  u xx subject to u  0, t   0, u 1, t   0 and
ww
u  x,0   sin  x,0  x  1.

Solution : Given Here 0  x  1, a  1, take h  0.2
a 1 2
k  h 2  k   0.2   k  0.02
2 2 . Form table.
HW (8) : Solve t u  u xx , given u  0, t   0, u  4, t   0 and u  x, 0   x  4  x 
assuming h = k = 1. Find the values of u upto t = 5.

 
.
1
Here   [Otherwise, the values of h & k are given]
Eg(9): 2
Solve ut  u xx ,0  x  1, t  0, u  x, 0   100  x  x  ,
2
u  0, t   u 1, t   0, 1 1
h  ,k 
4 16 .
1 1
Given ut  u xx , a  1, h  , k  .
Solution: 4 16
k 1/16  1
 2   1    1  
Let ah 1 1/16   2
We need to use formula ui , j 1    ui 1, j  ui 1, j   1  2 ui , j  1
   
.in
When   1, 1  ui , j 1   ui 1, j  ui 1, j  ui , j 
ing
The values of ui , j are calculated as follows:
 x direction 
eer
x
0 0.25 0.5 0.75 1
t
0 0 18.75 25 18.75 0
gin
 1
t 0 6.25 12.5 6.25 0
16
dir.
2
En
 0 6.25 0 6.25 0
16
3
arn
0 -6.25 12.5 -6.25 0

16
4
0 18.75 – 25 18.75 0
Le
16
w.
Topic 3: Finite difference solution of one dimensional heat equation by

implicit method(Crank–Nicolson’s method)
ww
Crank–Nicolson’s formula is
1
ui , j 1   ui 1, j 1  ui 1, j 1  ui 1, j  ui 1, j 
4
u
Eg (10) : Solve t  u xx  x  5, t  0, given that u  x, 0   20, u  0, t   0,
,0
u  5, t  =100. Compute u for the time–step with h = 1 by
Crank–Nicolson’s method.
Solution : Given ut  u xx , a  1, h  1,

k 2
  2  1  k  ah 2  11  k  1
By Crank–Nicolson’s formula, ah
1
4
u
The values of i , j are calculated as follows:
 x direction 
x
0 1 2 3 4 5
t
 0 0 20 20 20 20 0
t
.in
1 0 u1 u2 u3 u4 0

1
ing
Use Crank–Nicolson’s formula 4 ,
we get
eer
1
u1   0  0  20  u2   4u1  20  u2  4u1  u2  20  1
4
1
u2   u1  20  20  u3   4u2  40  u1  u3  u1  4u2  u3  40   2 
gin
4
1
u3   u2  20  20  u4   4u3  40  u2  u4  u2  4u3  u4  40   3
En
4
1
u4   u3  20  100  100   4u4  220  u3  u3  4u4  220   4
arn
4
Solving (1) and (2), we get
15u2  4u3  180   5
Le

4u2  15u3  380   6
w.

209u3  6420  u3  30.72
ww
  4   30.72  4u4  220  u4  62.68
  3  u2  4  30.72   62.68  40  u2  20.2

 1  4u1  u2  20  u1  10.05
 The values of u are 10.05, 20.2,30.72,62.68.

Eg (11) : Solve by Crank–Nicolson’s method, the equation u xx  ut

subject to u  x,0   0, u  0, t   0 & u 1, t   t for two time steps.
1 1
Given ut  u xx , a  1, x  0 to x  1, take h  .  k  ah 2 
Solution : 4 16
1
4
The values of ui , j are calculated as follows:
 x direction 
h
0 1 2 3 4
.in
t
0 0 0 0 0 0

ing
1 1
t 0 u1 u2 u3
16 16
dir.
2 2
eer
 0 u4 u5 u5
16 16
1
gin
Use Crank–Nicolson’s formula 4 ,

1
we get u1   0  0  0  u2   4u1  u2  1
4
En
1
u2   u1  0  0  u3   4u2  u1  u3   2 
4
arn
1 1 1
u3   u2  0  0    4u3  u2    3
4 16  16
Le
u u 1  1
  2   u2   1  2    u 2 = =0.0045
 4 4 64  224
w.
1
 1  u1  =0.0011
896
ww
  3  u3  0.0168
1 1
u4   0  0  u2  u5   u4   0.0045  u5    4
4 4
1 1
u5   u4  u1  u3  u6   u5   u4  0.0011  0.0168  u6    5 
4 4
1 1 2 1 3
u6   u5  u2     u6   u5  0.0045     6
4 16 16  4 16 
 u4  0.005899, u5  0.01913, u6  0.05277
 The values of u are 0.0011,0.0045,0.0168,0.005899, 0.01913, 0.05277 .

CW(12):Solve
ut  u xx ,0  x  2, t  0, given that u  0, t   u  2, t   0, t  0
x
and u  x, 0   sin
, 0  x  2. Using x  h  0.5, t  0.25
2 for
2 steps by Crank–Nicolson’s implicit finite difference method.
Solution: The u values are 0.3864,0.5469,0.3867,0.2115, 0.2991, 0.2115 .
Topic 3: One dimensional wave equation

 2u  2u
One dimensional wave equation is  a2
t 2 x 2
.in
h
valid for k 
a
After filling given boundary conditions
ing
(i.e., after filling first column, first row, last column by given conditions)
To find unknown 2nd row values alone by

eer
sum of just previous extreme columns
2
gin
To find unknown 3 onwards by

rd
sum of just previous extreme columns  previous 2nd value in same column
En
Eg (13) : Solve ytt  y xx upto t = 0.5 with a spacing of 0.1 subject to

y  0, t   0, y 1, t   0, yt  x,0   0 & y  x,0   10  x 1  x  .
arn
Solution :
Given differential equation ytt  y xx is of one dim. wave equation form.
Le
 2u  2u
i.e., One dimensional wave equation is  a2
t 2 x 2
w.
h
valid for k 
a
ww
2 2
Here coefficient of y xx in ytt  a y xx is 1  a  1  a  1
Given k  t difference  0.1  k
h h
WKT, k   0.1   h  0.1
a 1

11 .
Now draw table as follows:

x
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
t
0
10. 10. 10. 10. 10. 10. 10. 10.
(I 0 10+0.1(0.9)=10.09 0
16 21 24 25 24 21 16 09
row)
0.1 1 10.16 10. 10. 10. 10. 10. 10. 10. 5.0
(II 0  5.08 0
2 15 20 23 24 23 20 15 8
row)
0.2
0+10.15 – 10.09 5.1 10. 10. 10. 10. 10. 5.1 0.0
(III 0 0
= 0.06 2 17 20 21 20 17 2 6
row)
.in
0+5.12 – 5.08 0.0 5.1 10. 10. 10. 5.1 0.0 0.0
0.3 0 0
= 0.04 8 2 15 16 15 2 8 4
0.4 0
0+0.08 – 0.06
= 0.02
0.0
4
0.0
6
5.0
8 ing
10.
09
5.0
8
0.0
6
0.0
4
0.0
2
0
eer
0+0.04 – 0.04 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.5 0 0
gin
=0 0 0 0 0 0 0 0 0
y y
( (
En
0 1
, ,
arn
t t
) )
= =
Le
0 0
← ←
w.
CW (14) : Solve numerically, 4u xx  utt with the boundary conditions

u  0, t   0, u  4, t   0 and initial conditions ut  x, 0   0 and
ww
u  x, 0   x  4  x  , taking h = 1(for 4 time steps).

 2u  2u
2
 ,
HW (15) : Approximate the solution to the wave equation x t 2
0  x  1, t  0, u  0, t   u 1, t   0, t  0, u  x, 0   sin 2 x, 0  x  1
u  x, 0 
and  0, 0  x  1 with x  0.25 and t  0.25.
t

Topic 4: Two dimensional Laplace equation

Two dimensional Laplace equation is
 2u  2u
 2u  0  u xx  u yy  0   0
x 2 y 2
Standard five – point formula(SFPF):
1
ui , j  ui 1, j  ui 1, j  ui , j 1  ui , j 1 
4
ui , j 1
.in
ing
ui , j
ui 1, j ui 1, j
eer
ui , j 1
gin
Diagonal five – point formula(DFPF):

1
En
ui , j  ui 1, j 1  ui 1, j 1  ui 1, j 1  ui 1, j 1 

4
ui 1, j 1 ui 1, j 1
arn
Le
ui , j
w.
ui 1, j 1 ui 1, j 1
ww
For 3  3 matrix form:

B1 B12 B11 B10
B2 u1 u2 B9
u3 u4
B3 B8
B4 B7
B5 B6
To find u1, u2 , u3 , u4 :

13 .
Assume ui  0, for any one i

Initially take u4  0  or  take u4  some finite value
Find Rough Values [Initial values (or) Starting values] :

u1   B1  B3  u4  B11  4  by DFPF 
u2  u1  u4  B9  B11  4  by SFPF 
u3  u1  B3  B5  u4  4  by SFPF 
u4  u2  u3  B6  B8  4  by SFPF 
First Iteration onwards (apply only SFPF)
The Liebmann's iterative   n 1 1   n   n 1  n  n 1
 ui , j  ui 1, j  ui 1, j  ui , j 1  ui , j 1 
.in
formula is  4
Stop procedure until two consecutive iteration values are same.
ing
n  
n 1 
i.e., stop if ui  ui , i with error < 0.1
For 4  4 matrix form:
eer
B1 B16 B15 B14 B13
u1 u2 u3
gin
B2 B12
u4 u5 u6
B3 B11
u7 u8 u9
En
B4 B10
arn
B5 B9
B6 B7 B8
To find ui , i  1, 2,...,9 :
Le
Assume ui  0, for any one i

Initially take u4  0  or  take u4  some finite value
w.

 by SFPF, 
u5   B3  B7  B11  B15  4 
ww

 extreme boundary values only 
u1   B1  B3  u5  B15  4  by DFPF 
u3   B15  u5  B11  B13  4  by DFPF 
u7   B3  B5  B7  u5  4  by DFPF 
u9  u5  B7  B9  B11  4  by DFPF 

u2  u1  u5  u3  B15  4  by SFPF 
u4  u1  B3  u7  u5  4  by SFPF 
u6  u5  u9  B11  u3  4  by SFPF 
u8  u7  B7  u9  u5  4  by SFPF 

formula is  4
Stop procedure until two consecutive iteration values are same.
.in
n  
n 1  
i.e., stop if ui  ui , i with error < 0.1
2
Eg (16) : Solve  u  0, the boundary conditions
ing
are given below (give
only 3 iterations)
10 20
0 30
eer
u3 u4
20 40
u1 u2
40 50
gin
60 60 60 60
Solution :
En
Assume u4  0
arn
u1   20  0  60  60 4  35  by DFPF 
 
u2  35  0  50  60 4  36.25  by SFPF 
   by SFPF 
Le
u3  10  20  35  0 4  16.25
u4  16.25  20  40  36.25 4  28.125  by SFPF 
w.

ww
formula is  4
I iteration :
 
u 1   40  60  16.25  36.25 4  38.125
1
 by SFPF 
1
u
2
  60  50  38.125  28.125 4  44.0625  by SFPF 
1
u
3
  20  10  38.125  28.125 4  24.0625  by SFPF 
1
u4   20  40  24.0625  44.0625 4  32.0313  by SFPF 
II iteration :

15 .
 2
u
1
  60  40  24.0625  44.0625 4  42.0313  by SFPF 
 2
u
2
  60  50  42.0313  32.0313 4  46.0157  by SFPF 
 2
u
3
  20  10  42.0313  32.0313 4  26.0157  by SFPF 
 2
u4   20  40  26.0157  46.0157  4  33.0079  by SFPF 
III iteration
 3
u
1
  60  40  46.0157  26.0157  4  43.0079  by SFPF 
 3
u   60  50  43.0079  33.0079 4  46.5040  by SFPF 
.in
2
 3
u
3
  20  10  43.0079  33.0079 4  26.5040  by SFPF 
ing
 3
u4  20  40  46.5040  26.5040 4  33.252  by SFPF 
CW (17) : Obtain a finite difference scheme to solve the Laplace equation
eer
2
solve  u  0 at the pivotal points in the square mesh. Use
Liebmann’s iterative procedure.
1000 1000
gin
1000 1000
u1 u2
2000 500
u3 u4
En
2000 0
1000 500 0 0
arn
Solution : u1  1208.3, u2  791.7, u3  1041.7, u4  458.4

Le
2
HW (18) : By iteration method, solve the Laplace equation  u  0 over
the square region, satisfying the boundary conditions
u  0, y   0, 0  y  3, u  3, y   9  y, 0  y  3,
w.
u  x, 0   3 x, 0  x  3, u  x, 3  4 x, 0  x  3.
ww
Solution : u1  3.67, u2  7.34, u3  3.34, u4  6.67
Eg (19) : Find by Liebmann’s method the values at the interior lattice

points of a square region of the harmonic function u whose
boundary values are as shown.
11.1 17.0 19.7
0 18.6
u1 u2 u3
0 21.9
u4 u5 u6

0 u7 u8 u9 21.0
0 17.0
0 8.7 12.1 12.8 9.0
Solution :
u5   0  17.0  21.0  12.1 4  12.5  by SFPF, ext. bound. vals. only 
.in
u1   0  0  12.5  17.0 4  7.4  by DFPF 
u3  12.5  18.6  17.0  1.0 4  17.3  by DFPF 
 
ing
u7  12.5  0  0  12.1 4  6.2  by DFPF 
u9  12.5  9.0  12.1  21.0 4  13.7  by DFPF 
eer
u2  17.0  12.5  7.5  17.3 4  13.6  by SFPF 
u4   7.4  6.2  0  12.5 4  6.5  by SFPF 
gin
u6  12.5  21.0  17.3  13.7  4  16.1  by SFPF 

u8  12.5  12.1  6.2  13.7  4  11.1  by SFPF 
En

formula is  4
arn
I iteration :
  1
u 1  7.8  by SFPF  u41  6.6  by SFPF  u 7  6.6  by SFPF 
Le
1
1
u
2
 13.7  by SFPF  u 51  11.9  by SFPF  u81  11  by SFPF 
w.
1
u
1
 17.9  by SFPF  u 61  16.1  by SFPF  u9  14.3  by SFPF 
3
ww
II iteration :
 2  2
u
1
 7.9  by SFPF  u4 2  6.6  by SFPF  u 7  6.6  by SFPF 
 2
u
2
 13.7  by SFPF  u 52  11.9  by SFPF  u8 2  11.2  by SFPF 
 2
u
 2
3
 17.9  by SFPF  u 62  16.3  by SFPF  u9  14.3  by SFPF 
III iteration : Same as II iteration values. So stop the procedure.
Therefore, the solution of the given Laplace equation is
u1  7.9, u2  13.7, u3  17.9, u4  6.6, u5  11.9,
u6  16.3, u7  6.6, u8  11.2, u9  14.3

17 .
CW (20) : Solve
2 2
 u  u 1
2
 2  0, x  1, y  1 with h  and u  x, 1  x 2 ,
x y 2
u  1, y   y 2 .
Solution : u1  0.9, u2  0.13, u3  0.19, u4  0.13, u5  0.13,
u6  0.13, u7  0.19, u8  0.13, u9  0.19
 2u  2u
2
  0, 0  x, y  1 with u  0, y   10  u 1, y  and
HW (21) : Solve x y 2
.in
u  x, 0   20  u  x,1 . Take h  0.25 and apply Liebmann’s
ing
method to 3 decimal places.
Solution : u1  14.999, u2  16.249, u3  14.999, u4  13.749, u5  14.999,
u6  13.749, u7  14.999, u8  16.249, u9  14.999
eer
HW (22) : Solve the Laplace equation at the interior poins of the square
region given below.
1000 1000 1000
gin
500 500
u1 u2 u3
0 0
u4 u5 u6
En
0 0
u7 u8 u9
arn
0 0
0 0
0 0 0
Le
Solution : u1  71.5, u2  98.3, u3  71.5, u4  187.6, u5  250.1,

u6  187.6, u7  428.6, u8  526.8, u9  428.6
w.
Topic 5: Two dimensional Poisson equation

ww
Two dimensional Poisson equation is

 2u  2u
 2u  f  x, y   u xx  u yy  f  x, y    f  x, y 
x 2 y 2
x  ih, y  jk  jh  here 
Standard five – point formula(SFPF) for Poisson equation :
ui 1, j  ui 1, j  ui , j 1  ui , j 1  4ui , j  h 2  f  ih, jk 
Eg (23) : Solve the Poisson equation  u  10  x  y  10  , over the

2 2 2
square with sides x  0, y  0, x  3, y  3 with u  0 on the

boundary, taking h  1.

Y(X=0) Y=3
j=3 u=0
u3 (A) u4 (B) X=3
u=0 j=2
u1 (C) u2 (D) u=0
j=1
X(Y=0)
(0,0) i=1 i=2 i=3
i=0, j=0
Solution :
Standard five – point formula(SFPF) for Poisson equation :
ui 1, j  ui 1, j  ui , j 1  ui , j 1  4ui , j  h 2  f  ih, jh 
.in
 10  ih    jh   10 
2 2
 
ing  
 2
 10 i  j  102  ∵ h  1

At A  i  1, j  2  :
eer
u0,2  u2,2  u1,1  u1,3  4u1,2  10 12  22  10   150
 
gin
0+u 2  u3  0  4u1  150

u1  150  u2  u3  / 4
En
At B  i  2, j  2  :
u1,2  u3,2  u2,1  u2,3  4u2,2  10  22  22  10   180

arn
 
u1  0  u4  0  4u2  180
u2  180  u1  u4  / 4
Le
At C  i  1, j  1 : 0  u1  0  u4  4u3  10 12  120

u3  120  u1  u4  / 4
w.
At D  i  2, j  1 : u3  0  u2  0  4u4  10 15  150

ww
u4  150  u2  u3  / 4
Here u1  u4 , u2 , u3 .
Use any method to find u1 , u2 , u3 , u4 .
Now we are going to use Gauss – Seidel method:
Initially find u1 by putting u2  u3  0
Iterations becomes

19 .
1 1 1

u1  37.5 u2  63.75 u3  48.5
u1   65  2
u    60
2 2
u2  80 3
 3  3
u3   65
3
u1  70 u2  80
u1   75  4  4
4
u2  82.5 u3  67.5
u1   75  5  5
5
u2  82.5 u3  67.5
 u1  u4  75, u2  82.5, u3  67.5
.in
2
CW (24) : Solve the Poisson equation  u  81xy, 0  x  1, 0  y  1,
given that
ing
1
u  0, y   0, u  x, 0   0, u 1, y   100, u  x,1  100 & h  .
3
eer
Solution :  u1  u 4  51.0, u 2  76.5, u 3  25.8
2 2 2 u0
HW (25) : Solve  u  8 x y in the square mesh given
gin
on the four
boundaries dividing the square into 16 sub squares of length 1
unit.
En
Solution :  u1  u3  u7  u9  3, u2  u4  u6  u8  2
PART – A
arn
1. Write down the explicit finite difference method for solving one
dimensional wave equation.
Solution : The formula to solve numerically the wave equation
Le
a 2u xx  utt  0 is
   
w.
ui , j 1  2 1   2 a 2 ui , j   2 a 2 ui 1, j  ui 1, j  ui , j 1
2. Write down the standard five point formula to find the numerical
ww
solution of Laplace equation.

2
Solution : Standard five – point formula(SFPF) for    0 is
1
ui , j  ui 1, j  ui 1, j  ui , j 1  ui , j 1 
4
3. What is the error for solving Laplace and Poisson’s equations by finite
difference method?
Solution : For SFPF
h 4   4u  4u 
T.E.      ...
12  x 4 y 4 
i, j

For DFPF
h 4   4u  4u  4u 
T.E.   4  6 2 2  4   ...
6  x x y y  i , j
4. Write down the Crank – Nicolson’s formula to solve parabolic equation.
Solution :
1 1 1 1
 ui 1, j 1   ui 1, j 1     1 ui , j 1    ui 1, j   ui 1, j     1 ui , j
2 2 2 2
  
i.e.,  ui 1, j 1  ui 1, j 1  2    1 ui , j 1  2    1 ui , j   ui 1, j  ui 1, j 
Classify the partial differential equation u xx  2u xy  4u yy  0, x, y  0.
.in
5.
Solution : Given u xx  2u xy  4u yy  0, x, y  0.
ing
Here A  1, B  2, C  4
B 2  4 AC  12  0
 The given equation is elliptic.
eer
6. In one dimensional wave equation, write down the equation of explicit
scheme.
gin
Solution : The general form of the difference equation to solve the equation
utt  a 2u xx is
   
ui , j 1  2 1   2 a 2 ui , j   2 a 2 ui 1, j  ui 1, j  ui , j 1  1
En
If  a  1, the coefficient of ui , j in (1) is = 0

2 2
arn
The recurrence equation (1) takes the simplified form

ui , j 1  ui 1, j  ui 1, j  ui , j 1
Le
7. Classify the PDE y xx  xu yy  0.

Solution : Given y xx  xu yy  0.
w.
Here A  1, B  0, C   x
B 2  4 AC  4 x
ww
 The given equation is elliptic if x  0

parabolic x  0
hyperbolic x  0

21 .
Unit V Assignment problems:
.in
1. Compute for 4 time steps : ut  u xx :
ing
0  x  1, t  0 u ( x, 0)  100sin  x, u (0, t )  u (1, t ) choose h 1/ 4
u  2u
Solve the initial boundary value problem :  ,
t x 2
eer
2.
x
u ( x, 0)  cos ,  1  x  1, t  0 , u (1, t )  u (1, t )  0 for t  0
2
gin
3. UsingCrank-Nicolson's implicit scheme, Solve 16ut  u xx ,

0  x  1, t  0, given that u ( x, 0)  0, u (0, t )  0, u (1, t )  100t
En
u  2u
arn
Solve by Crank-Nicolson's scheme : 32  ,

4. t x 2
 t 
0  x  1, t  0 u ( x, 0)  0 , u (0, t )  100 sin   , u (1, t )  0
Le
 6 
 u  u 0  x  1, t  0 ;
2 2
w.
Solve 2  2 :
5. t x
ww
u ( x, 0)  100( x  x 2 ) ut ( x, 0)  0, u (0, t )  0, u (1, t )  0
6. Solve the Laplace equation over the square mesh of side 4 units,
satisfying the boundary conditions:
u(0, y) = 0, 0 ≤ y ≤ 4 ; u(4, y) = 12 + y , 0 ≤ y ≤ 4 ;
u(x, 0) = 3x, 0 ≤ x ≤ 4 ; u(x, 4) = x2 , 0 ≤ x ≤ 4.
7. Obtain the values at the interior points for the harmonic function
satisfying the following boundary conditions (4 iterations) after
deriving a suitable numerical scheme.
10 20 30

0 u1 u2 u3 40
0 u4 u5 u6 50
0 u7 u8 u9 60
0 70
50 60 70
0 80
Solve the partial differential equation xx 

u  u yy  10 x 2  y 2  10 
.in
8.
over a square with sides x = 0, y = 0, x = 3, y = 3, with the boundary
conditions u = 0 on all the boundary sides taking the mesh size h = 1.
ing
eer
gin
En
arn
Le
w.
ww

MA8491 Numerical Methods 01 - by LearnEngineering - in

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

MA8491 Numerical Methods 01 - by LearnEngineering - in

Uploaded by

Copyright:

Available Formats

For More Visit : www.LearnEngineering.

UNIT I : Solution of Equations and Eigenvalue Problems

Topic 1 : Solution of equation

f(x) < 0 for x1 < x < x2 and x3 < x < x4

Topic 2 :Fixed point iteration: x = g(x) method

Write the given equation as f(x) = 0.

For More Visit : www.LearnEngineering.in

f  x  0    0    0   100  100  0  ' ' ve 

f  x  1  1  1  100  98  0  ' ' ve 

f  x  2    2    2   100  88  0  ' ' ve 

f  x  4    4    4   100  20  0  ' ' ve 

 f  x  has a root between 4 and 5.

This equation f  x   x  x  100  0 can be written as

For More Visit : www.LearnEngineering.in

Note: In above problem, the root is 4.33105(at 5th decimal)

For More Visit : www.LearnEngineering.in

Eg.4. Solve the equation 3 x  cos x  2  0 by fixed point iteration

i.e, f  x  has a positive root.

d  2  cos x   1  d 1  sin x

For More Visit : www.LearnEngineering.in

CW.5. Solve the equationcos x  3 x  1 by iteration method upto 5 decimal.

 f  x  has a root between 0 and 1.

i.e, f  x  has a positive root.

This equation f  x   e  3 x  0 can be written as

For More Visit : www.LearnEngineering.in

Sol: Let f  x   2 x  log x  7  0

For More Visit : www.LearnEngineering.in

This equation f  x   2 x  log x  7  0 can be written as

For More Visit : www.LearnEngineering.in

Topic 3 : Newton’s method

Newton–Raphson method formula is

4. Derive Newton’s algorithm for finding the pth root of a number N.

5. Establish an iteration formula to find the reciprocal of a positive

For More Visit : www.LearnEngineering.in

For More Visit : www.LearnEngineering.in

x4  x3  x3  1.53039   1.52144

For More Visit : www.LearnEngineering.in

Let f  x   x tan x  1.28

1.04143 sec2 1.04143tan 1.04143

 0.93826  tan  0.93826 1.28

 A root of given equation is 0.93826

For More Visit : www.LearnEngineering.in

 A root of given equation is 0.35740

2 x  log10 x  7 by Newton–Raphson method.

N is a real number, hence evaluate 26 correct to 4 decimal places.

For More Visit : www.LearnEngineering.in

CW.20. Find the Newton–Raphson formula to find the value of N where

N is a real number, hence evaluate 3 17 correct to 5 decimal

Anna University Questions:

(Anna Univ. Nov./Dec, 2009(ECE)) Ans: x = 0.04347

For More Visit : www.LearnEngineering.in

Step2: Using elementary row operations reduce the given

Step3: By direct substitution we get the values for unknowns.

8. For solving a linear system AX = B, compare Gauss elimination

into upper triangular matrix into diagonal matrix or unit matrix

For More Visit : www.LearnEngineering.in

By back substitution to find the solution of the system of equations

For More Visit : www.LearnEngineering.in

For More Visit : www.LearnEngineering.in

AU5. Solve the given system of equations by Gauss elimination method:

x  from the first equation.

 (1) becomes a11