[go: up one dir, main page]

0% found this document useful (0 votes)
44 views5 pages

Range - Kutta Method

The document discusses numerical methods for solving ordinary differential equations, focusing on Euler's modified method and the Runge-Kutta methods. It explains how these methods improve accuracy in approximating solutions and provides specific examples and calculations. The text also outlines the derivation and application of first, second, third, and fourth order Runge-Kutta methods.

Uploaded by

pisot42384
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
44 views5 pages

Range - Kutta Method

The document discusses numerical methods for solving ordinary differential equations, focusing on Euler's modified method and the Runge-Kutta methods. It explains how these methods improve accuracy in approximating solutions and provides specific examples and calculations. The text also outlines the derivation and application of first, second, third, and fourth order Runge-Kutta methods.

Uploaded by

pisot42384
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

<br>

355
Mumerical Solution of Ordinary Differential Equations

y(0-2) in two steps from =+y, given


10. Use Euler's modified method to determine dx
=
that y(0) 1.
y =x+y and y = at x= 0. Find an approximate value
of y at x = 0-5, by
11. Give that
1l

dx
modified Euler's method.
by
=x+), y(0) =1. Find y for
x = 0-5 and 0-1 correct to 4 decimal places
12. Give that dy
dx
modified Euler's method.
equation S=-y"m, y = 2 at x = 0, by modified Euler's method
13. Solve the differential dx
two stages of 0-1 each.
and obtain atx=0-2 in
y
=
dy Vy and y
=1 when x= 1. Find approximate value ofy at x 2 in
14. Given that =2+
dx
steps of0-2, using Eulers modified method.
ANSWERS
2.2, 2-0048 3. 1-1448
= y(0-3) = 1-573
1. y(0-1)= 1:222, y(0-2) 1-375, 7.0-10046
5.0-0611 6.1-824
4. 3-18 11. 2-2352
9.2-5351, 2-6531 10. 1-2359
8. 4-0 14. 5-051.
13. 1-9227
12. 1-0520. 1-1104

9.6.RUNGE-KUTTA METHOD
As already discussed, Euler's method is less
efficient in practical problems
reasonable accuracy.
because it requires the step size h tobe small for computing
Also, the Taylor's series method of solving differential
equations requires much
However, there are some
labour in calculating higher derivatives numerically.
methods discovered by Carl Runge the German mathematician not
and developed by
require the
Wilhelm Kutta, known as Runge-Kutta methods whichor
do
calculations of higher accuracy. They
possess the advantage requiring only the
on the subintervals. The methods agree
function values at some selected points
r changes from
with Taylor's series solution upto the term h' where
method to method which is called the order of that method.
RungeKuttá Method. Consider the differential equation
(i) First Order
dy =fr, y) with y(*) =Y0
dx
Now by Euler's method V1=Yo th feo Yo)
=Yo +hyo (:: o=flxo Yo)l
+ h)
Using Taylor's series, '1=Yo
h2 .
=
y(x¢) + hy (xo) +)t

=Yo +
hyo+ +
2
Clearly, Euler's method agrees with Taylor's series solution upto the term
in h.

Hence, the Euler's method is the first order Runge-Kutta method.


<br>

Numerical
356 Analyt

Runge-Kutta Method. By Euler's modified method,


s
(ii) Second Order
have
+fug + h, yo
+ hfo))
Yo)
Y1=Y0 +,fro
Ifo +fig + h, y t hfo))
2;h
= yo

Using Taylor'sseries, we have


y1 =y(r + h)

=y(zo) + h,
() +5)+
Again, by Taylor's series
+ terms
fro + h, yo + hfo) =fxo, Yo) +h
cortaining second and higher powers ofA
+ 0(h?).
-fo+h| +

Putting thisexpression in (1), we get


1

+ 0(23)
=Yo + hfo + +fo
=Yo + hfo +
2
Y1=Yo + hfo + [: y=fx, y), y=f,)
..3)
method
comparing (2) and (3), we conclude that the modified Euler's
On Euler's method is the second
agrees upto the second order in h. Hence, the modified
order Runge-Kutta method.
=
If we now set h, hfo
|ky hflo +h, + k)
= yo
is
then second order Runge-Kutta formula
1
discussing
Method. Before
Runge-Kutta
(iii) Third order we first explain Runge-Kutta method
(whica
Runge-Kutta method of third order,
is not in course). and
point to + h between
we consider the middle
In Runge-Kutta method, 2

Xo +h and find the values off(x, y) ie., .at these points.


<br>

357
of Ordinary Diferential Equations
Numerical Solution

y corresponding to xo + is given by
The value of

the value of 'y at o + h is given by


d Yo).
I1=Y,+h=Yo +hfo
Yo) the values of fx, y) at
Thus the value of f(x, y) at (o, Yo)=fo,
= + Yx,+h/2)
(o + h/2 Jx, h/2) 1(*0 h/2
+

(u +h, yj) flxo + h, y;).


=
andthe values of f(, y)at
equation
Now integrating the differential
dy = f(x, y).
dx
yo + k), we get
From (x0, Yo) to (r0 +h,

dy = fx, y)dx

+*
y)dx
Or
fr,

By Simpson's rulel
which gives the value
of y
for equispaced,
Thus y = yo +k is a Runge formula
points.
can be done that Runge method agrees
We now discuss the R-K methods, it
with the Taylor's series solution
upto the order three in h.
Runge-Kutta method of order three.
Therefore, as such Runge method is the
If wenow set k= hflro, o)
h
2
hg = hfo + h, yo t k')
k'=hfu0 +h, yo k).
+
where
Then Runge-Kutta third order formula is
1
hy +4hy hg).
+
Y1 =Yo+
Fourth Order Runge-Kutta Method. As discussed above the first.
iv) methods, we notice that
Second and third order Runge-Kutta methods. In these
upto the terms ofh, h and h° respectively.
hey agree with Taylor's series solution Runge-Kutta method will agree with
nerefore, we can say that the fourth order
Taylor's
series solution upto the term h.
..(1)
Now we get ky=fro, yo) =fo
yo + gëy) ...(2)
kg = hfro + ao h,
<br>

358 Numerical
| Analysia

Nun

and k4 = hfuo + ayh, yo + agkg)


..4)
and
then fourth order formula is defined by
Y1 w,k + w,kg + wgkg +
= Yo + wgk4

Expanding y, by Taylor's series, we have


Y1=y(*0 +h)
=
y(*o) + hy(ro) + +y)
iu
iv

325tai )+ 0(4

+ 0(h).
Or Y1 =Yo + hyo
21o Yo+hß ..(6)

dy =f, y)
ie., y' =fso yo =fo
Since
dx
Now +yf+ffy
dy
y"=feu t2fiy +f"fyy +fgfy +fi,y
y=fery t 3fy + 3ffryy +fy fece+ 2fy +ffy)

For convenience, assuming

and sQg =fe + 3ffcey + 3f syy +f°gyi agul net &ro

=03 +f, 2 + 301 y tiy) +Iy


and 1ofoz sa
o= o= 20 +y,
10 O10: not

p2

Putting these values in (7), we get and


..
y1=Y0 + hfo +110 +3 (20 +fy, 1o) t
+
4!
(o30 +fy 20 + 3010 iy, tfo fyy)+f 2
Again expanding R.H.S. of (2), (3) and (4) by Taylor's series and ustn
convenient notation, we get
1

. 1

and
1
<br>

Solution of
Ordinary Differential Equations
Naumerical
359

and k4

Putting these values in (5), we


get.

y=0 + (w1 + Wg
tWg + W4) hfo + (wgc6 + wga+ wc) hon
(w_o + wgaí + wa) O20 + h° (aoajwg + aj0glU4) y, O1o
2

Comparing (8) and (9), we get

Wj +
W+Wg + w4 =1
81e1 wg0gaj + wg4, & = 1

2
1 1
12

1
4 24
so they can be solved but it is
There are eight equations in seven unknowns,
not an easy task, so if we set

1 1

2
and W1= W4 =
3
Then fourth order formula reduces to

h
2

k4 = hf(xo + h, o + k)
and
+ 1
+ k4).
1=Yo (k + 2kg + 2kg
This leads method.
to
fourth order Runge-Kutta

You might also like