Department of Mathematics

UNIT - III
NUMERICAL METHODS

The study of numerical analysis is aimed at providing convenient methods for obtaining useful
solutions to problems of advanced learning in science and technology. Besides this, analytical
solutions to a large number of problems are not available. Numerical analysis provides methods
for obtaining approximate solutions to such problems also. Numerical methods, in general, are of
repeated nature. In each step, better approximation to the exact solution of the problem is
obtained and the process is continued till accuracy to the desired degree is arrived at. In this unit
some numerical methods to solve algebraic, transcendental and ordinary differential equations
are discussed.

Algebraic and Transcendental Equation:

n n−1 n−2
An expression of the form f ( x )=a 0 x +a1 x +a2 x +.. .+a n−1 x+an , where ai (i = 0, 1, 2,
3…) are constants (a 0≠0) and n is a positive integer, is called a polynomial in x of degree n.
The polynomial f (x) =0 is called an algebraic equation of degree n.

Examples:
2
(1) 3 x −6 x +9=0

(2) (2 x−8 )(x +5 )=0

If f ( x ) contains some other functions such as trigonometric, logarithmic, exponential etc., then
f ( x )=0 is called a transcendental equation.

Examples:
x
(1) 2 e −4=0
x
(2) cos x−xe =0

Root of an equation: The value ‘a’ of x which satisfies f ( x )=0 is called a root of f ( x )=0 .
Geometrically, a root of f ( x )=0 is that value of x where the graph of y=f ( x ) crosses the
x-axis.

1
Second Semester Vector Calculus, Laplace Transforms and Numerical Methods (22MA21A)
 Department of Mathematics

'
Simple root: A number ξ is a simple root of f ( x )=0 if f ( ξ )=0 , f ( ξ )≠0 .

Multiple root: A numberξ is a multiple root of multiplicity m if

f (ξ )=f '(ξ )=.. .. .=f (m−1) (ξ ) ∋ f ( m )(ξ )≠0 .
m
i.e., f ( x )=(x −ξ ) g ( x ) ; g (ξ )≠0.
The process of finding the roots of an equation is called solution of that equation. This is a
problem of great importance in Scientific and Engineering studies. To solve higher order and
transcendental equations no analytical methods exist. Such equations are solved by numerical
methods.

Basic properties of equations:

1. If f ( x ) is exactly divisible by x-a, then ‘a’ is a root of f ( x )=0 .

2. Every polynomial equation of the nth degree has only 'n' roots (real or imaginary).
3. In an equation with real coefficients, imaginary roots occur in conjugate pairs, i e., if is a
root of the equation f ( x )=0 , then must also be its root.
4. Every polynomial equation of odd degree has atleast one real root.

Intermediate value property: If f ( x ) is continuous in the interval [a, b] and f (a ) , f (b ) have

different signs, then the equation f ( x )=0 has at least one root between x = a and x = b.

Example:

1. Solve the equation , one root being .

Solution:
Since one root is , the other root must be
Therefore, the factors corresponding to these roots are and
or = =
3 2
is a divisor of 3 x −4 x + x +88=0
3x+8 = 0
x = - 8/3

Initial approximation:

2
Second Semester Vector Calculus, Laplace Transforms and Numerical Methods (22MA21A)
 Department of Mathematics

The common method used to obtain the initial approximation to the root is based upon the
intermediate value theorem which states that “If f(x) is a continuous function in the closed
interval [a, b] and if f(a) and f(b) are of opposite signs [i.e., f(a).f(b)<0] then the equation f(x) = 0
has atleast one simple root (or odd number of roots) lying between a and b”. By using this
property, we take a (or) b as the initial approximation of the root to start the Iterative process.
Regula-Falsi method (Method of false position):

This is method of finding the real root of an equation f ( x )=0 . Here we choose two points
such that and are of opposite signs i.e., the graph of y=f(x) crosses the x- axis
between these points a root lies between 0
x and x
1 and consequently f ( x 0 ) f ( x1 )<0 .
Consider the equation f(x) = 0, where f(x) may be Algebraic (or) Transcendental which is known
to have a root between [x0 , x1 ] then f ( x 0 ) f ( x1 )<0 .

Approximate the function f(x) by the straight line joining the points [x 0 , f(x0 )] and [x1 , f(x1 )].
Let this straight line cut the x-axis at x2 .
The equation of the chord joining the two points [x0 , f(x0 )] and [x1 , f(x1 )] is given by,
y−f ( x 0 ) x−x 0 f ( x 1 )−f ( x 0 )
= ⇒ y−f 0= ( x −x 0 ) ---------( 1)
f ( x 1 )−f ( x 0 ) x 1−x 0 x 1 −x0
The method consists in replacing the curve AB by means of the chord AB and taking the point of
intersection of the chord with x - axis as an initial approximation to the root. The point of
intersection is obtained by putting y = 0
f ( x 1 )−f ( x 0 )
0−f (x 0 )= (x 2 −x 0 )
∴ Equation (1) becomes, x 1− x0

f ( x 0 )( x 1−x 0 )
⇒ x 2 −x 0 =−
f ( x 1 )−f ( x 0 )

3
Second Semester Vector Calculus, Laplace Transforms and Numerical Methods (22MA21A)
 Department of Mathematics

( x 1 −x 0 )
⇒ x 2 =x 0 − f ( x0 )
f ( x 1 )−f ( x 0 )
x 0 f ( x 1 )−x1 f ( x 0 )
x 2= ----(2 )
f ( x1 )−f ( x 0 )

The formula given (2) is the first approximation to the root.

If f (x2 ) and f (x0 ) are opposite signs then replace x1 by x2 and draw a straight line joining f (x2 )
and f (x0 ) to find the new intersection point. If f (x 2 ) and f (x0 ) are of same sign then x0 is
replaced by x2 and repeat the above procedure till the root is found to the desired accuracy.
Repeating the above process n – times, we get the general formula given by:

x n−1 f ( x n )−x n f ( x n−1 )

x n+1= ---------(3) n=1,2,3,. .. .
f ( x n )−f (x n−1 )
Note:
1. Regula - Falsi method has linear rate of convergence.
2. Geometrically, this method is equivalent to replacing the curve y = f(x) by a chord that passes
through the points A [a, f (a)] and B [b, f (b)] and intersection of chord with x-axis is obtained as
solution of equation.

Examples:
1. Find a real root of the equation by the method of false position correct to three
decimal places.
Solution: Let f(x) =
f (1) = -6 (-ve), f (2) = -1 (-ve), f(3) = 16 (+ve)
A root lies between 2 and 3
therefore x 0 =2 x1 =3 f ( x 0 )=−1 f ( x 1 )=16
By the method of false position,

4
Second Semester Vector Calculus, Laplace Transforms and Numerical Methods (22MA21A)
 Department of Mathematics

1
x 2 =2+ =2. 0588
We get, 17

Since f (2. 0588 )=−0 . 3908(−ve) , therefore the root lies between 2.0588 and 3.

Now, x 0 =2. 0588 , x1 =3 , f ( x 0 )=−0 .3908 , f ( x 1 )=16

0 . 9412
x 3 =2. 0588− (−0. 3908 )=2. 0813
16 .3908
Repeating this process, the successive approximations are

Hence, the root is 2.094 correct to 3 decimal places.

2. Find the root of the equation using Regula - Falsi method correct to four decimal
places.
Solution:
x
Let f ( x )=cos x −xe =0
f (0) = 1 (+ve), f (1) = -2.17798 (-ve)
Root lies between 0 and 1.
therefore, x 0 =0 , x 1=1 , f (x 0 )=1 , f ( x 1 )=−2. 17798
By the method of false position,

Since f (0.31467) = 0.51987

The root lies between 0.31467 and 1

therefore x 0 =0 .31467 , x 1 =1 , f ( x 0 )=0. 51987 , f ( x1 )=−2. 17798

0 .68533
x 3 =0 .31467− (0 . 51987)=0 . 44673
2. 69785

5
Second Semester Vector Calculus, Laplace Transforms and Numerical Methods (22MA21A)
 Department of Mathematics

The successive approximations are

Hence the root is 0.5177.

3. Find a real root of the equation by Regular - Falsi method correct to four decimal
places.

Solution: Let
f (1)=−ve , f (2)=−ve and f (3 )=+ ve
Therefore, a root lies between 2 and 3
Now, x 0 =2 , x 1 =3 , f ( x 0 )=−0 .59794 , f ( x 1 )=0 . 23136
By Regular-Falsi method

= 2.72102

⇒ the root lies between 2.72102 and 3

0 . 27898
x 3 =2. 72102+ (0 . 01709)=2. 74021
0. 23136+0. 01709

Repeating this process, the successive approximations are

Hence the root is 2.7406 correct to 4 decimal places.

Newton-Raphson method:

Let f(x) = 0 ---------- (1), be the given equation, where f(x) may be Algebraic (or) Transcendental
and x0 be the initial approximation to the root.
Let x1 = x0 + h ---------- (2) be the exact root of the equation f(x) = 0, where h is very small
quantity.

6
Second Semester Vector Calculus, Laplace Transforms and Numerical Methods (22MA21A)
 Department of Mathematics

From (1), we have f(x1 ) = f (x0 + h) = 0 ---------- (3)

Expanding f (x0 + h) by Taylor’s series expansion we get,
h h2 h3
f ( x 0 +h )=f ( x 0 )+ f '( x 0 )+ f ''( x 0 )+ f '''( x 0 )+. .. .
1! 2! 3! ------------(4)
Since f(x1 ) = f (x0 + h) = 0 and h is very small, so higher powers of h can be neglected.
∴ Equation (3) becomes,

f ( x 0 +h )=f ( x 0 )+hf ' ( x 0 )

⇒ 0=f ( x 0 )+hf '( x 0 ) (∵ f ( x 0 + h)=0)

f ( x0 )
⇒ h=−
f ' ( x0 )
----------------------(5)
Substituting (5) in (2), we get the first approximation to the root given by,
f ( x0 )
x 1=x 0 −
f ' ( x0 ) .

Similarly starting with x1 and repeating the above procedure we get the second approximation to
the root given by,
f ( x1 )
x 2 =x1 −
f ' ( x1 )
th
Proceeding like this, we get the n approximation to the root given by,

f ( xn )
x n+1=x n − ---------( 6 )
f ' ( xn )

The formula given in (6) is known as Newton-Raphson Iterative formula to find the root of the
equation f(x) = 0.

Note: 1 Newton’s formula converges provided initial approximation is chosen sufficiently

close to the root.
Note: 2 Newton’s method has a quadratic convergence. i.e., the subsequent error at each step is
proportional to the square of the previous error.

7
Second Semester Vector Calculus, Laplace Transforms and Numerical Methods (22MA21A)
 Department of Mathematics

Geometrical Interpretation:

Let be a point near the root of the equation

Then, the equation of the tangent at is

It cuts the x-axis at

If is a point corresponding to on the curve, then the tangent at will cut the x- axis at ,
which is nearer to is a second approximation to the root. Hence the method consists in

8
Second Semester Vector Calculus, Laplace Transforms and Numerical Methods (22MA21A)
 Department of Mathematics

replacing the part of the curve between the point and the x- axis by means of tangent to the
curve at .

Examples:
1. Apply Newton-Raphson’s method to obtain the smallest positive root of the equation
cos(x) - x ex = 0.
Solution: Given cos(x) - x ex = 0.
' x x
⇒ f(x) = cos(x) - x ex and f ( x )=−sinx−xe −e
Since f (0) = 1>0 and f (1) =-2.1780 <0, the root lies in the interval (0,1)
Let x0 = 1 be the initial approximation to the root.

The approximate root is calculated as follows:

k xk Δx k =
f ( xk ) x k +1=x k−Δx k f ( x k+1 )
f ' ( xk )
0 1.0 0.3469 0.65307940 -0.4606
1 0.65307940 0.1217 0.53134337 -0.4180
3 0.53134337 0.1343 0.51790991 -0.4641
4 0.51790991 0.1525 0.51775738 -0.5926
5 0.51775738 0.1948 0.51775736 -0.2910

Hence, x = 0.51775736 is the approximate root.

2. Find by Newton’s method, the real root of the equation .

Solution:
Let

So, a root of lies between 0 and 1.

It is nearer to 1 let us take

9
Second Semester Vector Calculus, Laplace Transforms and Numerical Methods (22MA21A)
 Department of Mathematics

Therefore, Newton’s iteration formula is

Put n = 0 first approximation is

Put n = 1 second approximation is

Hence the desired root is 0.6071 correct to 4 decimal places.

3. Using Newton’s iterative method, find the real root of correct to five decimal
places.
Solution:

Take

The approximate root is calculated as follows

10
Second Semester Vector Calculus, Laplace Transforms and Numerical Methods (22MA21A)
 Department of Mathematics

k x k +1=x k−Δx k
1 2.81
2 2.741
3 2.74064
4 2.74065

5 2.74065

Since
Hence the required root is 2.74065.

4. Evaluate to four decimal places by Newton’s iterative method.

Solution:
Let

Therefore

Newton’s iterative formula is

Since f (3) =-3, f (4) = 4

Root lies between 3 and 4

since up to 4 decimal places, therefore .

11
Second Semester Vector Calculus, Laplace Transforms and Numerical Methods (22MA21A)
 Department of Mathematics

Failure of Newton-Raphson method:

 Inflection points in vicinity of root.
 Oscillate around local maximum or minimum.
 Jump away for several roots.
 Disaster from zero slope.

Numerical solutions of ordinary differential equation:

th
A general solution of a differential equation of the n order has n arbitrary constants. In order to
compute Numerical solution of such an equation we need n conditions. If all the n conditions are
prescribed at the initial point (one point) only, then it is called initial value problem. If the n
conditions are specified at two (or) more points then it is called Boundary Value Problem.

Taylor’s series Method:

Consider the first order ordinary differential equation,
dy
=f ( x,y ), subject to initial conditions y( x 0 )= y 0
dx ----- (1)

Let y= y (x ) be the exact solution of equation (1) expanding y(x) in Taylor series about the
point x0 , we get,
( x−x 0 ) ( x−x 0 )2 ( x−x 0 )3
y ( x )= y ( x0 )+ y '( x 0 )+ y ''( x 0 )+ y '''( x 0 )+…
1! 2! 3!
' ''
Since y ( x 0 )= y 0 , y ( x 0 )= y , y ( x 0 )= y etc., the above equation becomes
' 0 '' 0

12
Second Semester Vector Calculus, Laplace Transforms and Numerical Methods (22MA21A)
 Department of Mathematics

( x−x 0 ) ( x−x 0 )2 (x −x 0 )3
y ( x )= y 0 + y '0 + y ''0 + y '''0 +…
1! 2! 3!
Setting h = x1 - x0 and denoting y(x) = y(x1 ) = y1 , the above equation becomes,
h h2 h3
y 1= y 0+ y' 0 + y ''0 + y'''0 +....
1! 2! 3! --------------(2)
This is the Taylor series solution at the point x 1 = x0 + h. Starting with this x1 and repeating the
above procedure we get the solution at the point x2 as,
th
Proceeding like this we get the Taylor series solution at the n step given by,
h h2 h3
y n+1 = y n + y ' n + y ''n + y'''n +.. ..
1! 2! 3! ---------------(3)
The formula given in (3) is known as Taylor’s series method.
Note: It is a single step method.

Example:
dy
=x 2 + y 2
1. Use Taylor’s series method to find y at the points x = 0.1 and x = 0.2, given that
1 2 dx ,
y (0) = 1.
Solution: Here x0 = 0, the Taylor’s series solution of the given problem is

x 2 '' x3 x4
y= y ( 0 )+xy ' ( 0 )+ y ( 0 )+ y ''' ( 0 )+ y iv ( 0 )+.. .. .
2! 3! 4! with y (0) = 1. (i)

y ' =x 2 + y 2 , y ''=2 x+ 2 yy ' , y''' =2+2 { yy '' + ( y ' ) } , y iv =2 { yy ''' +3 y ' y '' }
2
Also, and so on.

Using the condition y (0) = 1 in these we obtain,

y ' ( 0 )=1 , y '' ( 0 )=2 , y ''' ( 0 )=8 , y iv ( 0 )=28 .

Using these and the condition y (0) = 1 in (i) we get,

y=1+ x + ( x2
2!
×2 + )(
x3
3!
×8 +
x4
4! )(
×28 + .. .. . .. )
13
Second Semester Vector Calculus, Laplace Transforms and Numerical Methods (22MA21A)
 Department of Mathematics

4 7
=1+ x + x 2 + x 3 + x 4 +. . .. .. (ii )
3 6

Under the given condition ( 0 )

y x = y ( 0 )=1
, this is the Taylor’s series solution for y at a point x in a
neighbourhood of

x 0 =0
.

x=x 1 =0 .1 x 1=0. 1
For , expression (ii) yields the solution at as,

4 7
y ( 0 .1 ) =1+ ( 0 .1 ) + ( 0 . 1 )2 + ( 0 . 1 )3 + ( 0 . 1 )4 +.. . .. ..≈1 .11145
3 6

x=x 2 =0 .2 x 2 =0 .2
For , expression (ii) yields the solution at as,

4 7
y ( 0 .2 ) =1+ ( 0 .2 ) + ( 0 . 2 )2 + ( 0 . 2 )3 + ( 0 .2 ) 4 +. .. . .. .≈1 . 25253 .
3 6

Runge - Kutta fourth order method:

The general formula for finding the solution of the initial value problem
dy
=f ( x,y ),y (x 0 )= y 0
dx is given by:
1
6[ 1
y n+1 = y n + K +2K2 +2K3 + K 4 ]

Where K 1 =h f ( x n ,y n ) (
K 2 =h f x n +
h
2
K
,y n + 1
2 )
,

(
K 3 =h f x n +
h K
,y + 2
2 n 2 ) and K 4 =h f ( x n + h,y n + K 3 )

14
Second Semester Vector Calculus, Laplace Transforms and Numerical Methods (22MA21A)
 Department of Mathematics

Example:
' 2
1. Using fourth – order Runge – Kutta method, solve the problem y =x + y , y (0) = 1 at the
points x = 0.1 and x = 0.2 in steps of 0.1.
f ( x , y )=x + y 2 , x 0=0 , y 0 =1 x =x +h=0 .1
Solution: Here, , h=0.1, so that 1 0 and
x 2 =x1 + h=0. 2
y 1 = y ( x 1 )= y ( 0 . 1 ) and y 2 = y ( x 2 ) = y ( 0 .2 )
Step 1: We have to find . Hence,
k 1=hf ( x 0 , y 0 ) =( 0 .1 )×f ( 0 , 1 )=( 0 . 1 )×( 0+ 1 )=0. 1 ,
2

( h k
)
k 2=hf x 0 + , y 0 + 1 =( 0 .1 )×f ( 0 . 05 ,1 . 05 )
2 2
= ( 0. 1 )× {0 .05+ ( 1. 05 )2 }=0 . 11525

( h k
)
k 3=hf x 0 + , y 0 + 2 =( 0 . 1 )×f ( 0 . 05 , 1. 057625 )
2 2
= ( 0. 1 )× {0 .05+ ( 1. 057625 )2 }=0. 116857
k 4 =hf ( x 0 + h , y 0 +k 3 ) =( 0 .1 ) ×f ( 0 . 1, 1 .116857 )

= ( 0. 1 )× {0 .1+ ( 1 .116857 )2 }=0 . 13474

1
k= ( k +2 k 2 +2 k 3 + k 4 )
6 1
and,
1
= ( 0 .1+2×0 .11525 +2×0 . 116857+0. 13474 )=0 . 11649
6
x 1=0. 1
Thus, the solution at is,
y 1 = y 0 + k=1+0 . 11649≈1 .1165
k 1 , k 2 , k 3 ,k 4 and k
by changing (
x0 , y0 )
to (
x 1 , y 1)
Step 2: Again, we compute
k 1=hf ( x 1 , y 1 ) =( 0 . 1 )×f ( 0 . 1 , 1 .1165 )=0 . 134657 ,

h
( k
)
k 2=hf x 1 + , y 1 + 1 =( 0 . 1 )×f ( 0 .15 , 1. 1838 )=0. 15514
2 2
,

( h
k 3=hf x 1 + , y 1 +
2
k2
2 )
= ( 0. 1 )×f ( 0 .15 , 1 .1941 )=0. 15759 ,

k 4 =hf ( x 1 +h , y 1 + k 3 )=( 0 . 1 )×f ( 0. 2 , 1. 27409 )=0 . 18233 ,

1
k= ( k +2 k 2 +2 k 3 + k 4 )
6 1
and

15
Second Semester Vector Calculus, Laplace Transforms and Numerical Methods (22MA21A)
 Department of Mathematics

1
= ( 0 .134657+2×0 . 15514+2×0. 15759+0 . 18233 )=0. 15708
6 .
x 2 =0 .2 y 2 = y 1 + k=1 .1165+0 .15708=1 .27358
Hence, the solution at is, .

Predictor-Corrector Method:
The methods discussed above to solve a first order differential equation, require information
about the solution at a single point say xn for computing the solution at xn+1 . Hence these
methods are called single-step methods.
In the Predictor-Corrector Methods, a predictor formula is used to predict the value y i+1 of y at x
i+1 and then a corrector formula is used to improve the value of y i+1 .The corrector formula is used
repeatedly until two consecutive values of yi+1 are almost equal.

Milne’s Predictor-Corrector method:

The Predictor-Corrector methods form a large class of general methods for numerical integration
of ordinary differential equations. As an illustration, consider Milne's method for the first-order
equation y '( x )=f ( y ( x ), x ) initial value y(x0) = y0.
Define
y n = y ( x 0 + nh),
y ' n = y '( x 0 + nh)=f ( y n , x 0 +nh )
Then by Simpson's rule

y 'n+1 =f ( y n+1 , x +( n+1)h)

This corrector equation is an implicit equation for yn+1; if h is sufficiently small, and if a first
approximation for yn+1 can be found, the equation is solved simply by iteration, i.e., by repeated
evaluations of the right-hand side. To provide the first approximation for yn+1, an explicit
predictor formula is needed, e.g., Milne's formula
y n+1 = y n−3 +( 4 h/3 )(2 y ' n− y ' n−1 +2 y ' n−2 )+O( h5 )
Predictor Formula:
4h
y (n+1
p)
= y n−3 +
3
[ 2 y 1n−2 − y 1n−1 +2 y 1n ]
Corrector Formula:
h 1
y (n+1
c)
= y n−1 +
3
[ y n−1 +4 y 1n + y 1n+1 ]
.

16
Second Semester Vector Calculus, Laplace Transforms and Numerical Methods (22MA21A)
 Department of Mathematics

The need for a corrector formula arises because the predictor alone is numerically unstable; it
gives spurious solutions growing exponentially. Milne's predictor uses four previous values of y,
these values can be obtained by using Taylor’s or Runge-Kutta method.
Example:
1. Using Milne's Predictor-Corrector method evaluate the integral of
y '−4 y=0 at 0. 4 , given that y(0)=1, y(0 .1)=1.492, y(0 . 2)=2 .226, y(0.3)=3 . 320
Solution:
dy
=4 y
Here, dx So f ( x , y )= y '=4 y
x =0 , x 1=0 . 1 , x 2 =0 . 2, x 3 =0 . 3 , x 4 =0 . 4
h=0.1, 0
For predictor value of y (0.4):
4h
y (n+1
p)
= y n−3 +
3
[ 2 y 1n−2− y 1n−1 +2 y 1n] with n=3
4h
y (4p )= y 0 +
3
[ 2 y 11 − y 12 +2 y 13 ]
Here, h=0.1,
y 0 =1 , y 11=4 y 1 =4 (1. 492)=5 . 968
y 12 =4 y 2=4(2 . 226)=8 . 904
y 13 =4 y 3 =4 (3. 32)=13 .28
p 4(0 . 1)
∴ y 4 = y ( x 4 )= y (0 . 4 )=1+ [ 2(5 . 968)−8. 904 +2(13 .28 )]=4 . 9456
3
So y 14 =4 y 4 =4 (4 . 9456 )=19 .7824
For corrector value of y (0.4):
h 1
y (n+1
c)
= y n−1 +
3
[ y n−1 +4 y 1n + y n+1
p
]
h
y (4c )= y 2 + [ y 12 + 4 y 13 + y 14 ]
with n = 3, 3

0.1
∴ y 4 = y (0. 4 )=2. 226+ (2 .226 +4 (13 .28 )+19 .7824=4 . 95288
3 (First approximation of y4)

Second approximation of y4:

(C ) h 1
y4 1 = y2 +
3
[ y2 + 4 y3 + y4 ]
1 ( c)

y (1 )=4( 4 . 95288)=19 .81152

4

17
Second Semester Vector Calculus, Laplace Transforms and Numerical Methods (22MA21A)
 Department of Mathematics

0.1
∴ y 4 = y (0. 4 )=2. 226+ (2 .226 +4 (13 .28 )+19 .. 81152)= 4 .731251
3 .

Similarly, third approximation of y4:

0.1
∴ y 4 = y (0. 4 )=2. 226+ (2 .226 +4 (13 .28 )+4 (4 . 731251))=4 .701700
3 .

Fourth approximation of y4:

0.1
∴ y 4 = y (0. 4 )=2. 226+ (2 .226 +4 (13 .28 )+4 (4 . 701700 ))=4 . 69776
3 .

Fifth approximation of y4:

0.1
∴ y 4 = y (0. 4 )=2. 226+ (2 .226 +4 (13 .28 )+4 (4 . 69776 ))=4 . 697234
3 .
∴ The approximate value of y (0.4) is 4.697.

2. Using the Milne’s predictor-corrector method, find an approximate solution of the initial value

problem at the point x = 2, given y (1) = 2, y (1.25) = 3.13, y (1.5) = 4.5 and y
(1.75) = 6.13. Apply the corrector twice.

Solution: Here , x0 = 1, y0 = 2, x1 = 1.25, y1 = 3.13, x2 = 1.5, y2 = 4.5, x0 = 1, y3 = 6.13

h = 0.25, also f1 = f (x1, y1) = 5.008, f2 = f (x2, y2) = 6.00, f3 = f (x3, y3) = 7.0057

Using Milne predictor yields the value of y4 as: = 8.0091

This gives f4(p) = f (x4, y4) = 8.0091

The Milne corrector gives the improved value of y4 as:

= 8.0027,

Further computing f4(p) = f (x4, y4) = 8.0027, applying the corrector again y4(c) = 8.0021

The approximate solution for ‘y’ at the point x =2, is 8.0021.

18
Second Semester Vector Calculus, Laplace Transforms and Numerical Methods (22MA21A)
 Department of Mathematics

Exercise:

x
1. Determine the smallest positive real root of xe −2=0 to four significant figures using
false position method.
(Ans: 0.8526)

2. Solve the equation x tan x=−1 by Regular - Falsi method between 2.5 and 3 correct to
three decimal places.
(Ans: 2.7982)

3. Find the real root of the following equations by Newton - Raphson’s method
i) x3 – x – 11 = 0 in the interval (2, 3) (Ans: x = 1.532)
ii) x3 – 3x + 11 = 0 in the interval (1, 2) (Ans: x = 1.532)

4. The volume ‘V’ of a liquid in a spherical tank of radius ‘a’ is related to the depth ‘l’ of the liquid
given by l3 - 3l2 + 0.4775 = 0, using Newton-Raphson method find a positive real root of the
equation which lies between 0 and 1.
(Ans: 0.431137997)

5. If y '=1+xy and y = 2 when x = 0, use Taylor's series method to obtain the value of y for
x = 0.4.
(Ans: 2.5884139)

x
6. Solve the differential equation y '=2 y+3e with y (0) = 0 using Taylor's series method
and find the value of y for x = 0.1, 0.2. Compare your answer with exact values.

(Ans: Calculated values: y (0.1) = 0.3487, y (0.2) = 0.8112

Exact values: y (0.1) = 0.3486, y (0.2) = 0.8112)

dy
=log( x + y )
7. Solve the initial value problem dx when the initial condition that y = 1 when
x = 0 use Runge - Kutta method to find y (0.2) and y (0.5) in more accurate form.

19
Second Semester Vector Calculus, Laplace Transforms and Numerical Methods (22MA21A)
 Department of Mathematics

(Ans: y (0.2) = 1.0082, y (0.5) =1.0490)

2 2
8. Using Runge - Kutta 4th order method compute y (0.2) and y (0.4) from y '=x + y , y (0)
=1, taking h = 0.1.
(Ans: y (0.2) = 1.0207, y (0.4) = 1.038)
dy
=3 x−4 y
9. Solve the initial value problem dx , y (0) =2 at x =0.4 taking h = 0.2 by Runge-
Kutta method of order 4.

(Ans: y (0.4) = 0.5543)

dy
=1+ xy 2
10. Solve the initial value problem dx , y (0) = 1 for x = 0.4 by using Milne's
method when it is given that:
x: 0.1 0.2 0.3
y: 1.105 1.223 1.335
(Ans: y =1.538 at x = 0.4)

Video link:

https://www.youtube.com/watch?v=QwyjgmqbR9s

20
Second Semester Vector Calculus, Laplace Transforms and Numerical Methods (22MA21A)