Runge-Kutta Methods

Learning Objectives of Lesson 3

 To understand the motivation for using
Runge Kutta method and the basic idea
used in deriving them.
 To Familiarize with Taylor series for
functions of two variables.
 Use Runge Kutta of order 2 to solve
ODEs.

1
Motivation
 We seek accurate methods to solve ODEs
that do not require calculating high order
derivatives.
 The approach is to use a formula
involving unknown coefficients then
determine these coefficients to match as
many terms of the Taylor series
expansion.

General Form of Runge-Kutta Method

Masalah nilai awal: = 𝑓 𝑡, 𝑦 , 𝑡 > 0, 𝑦 0 = 𝑦 .

Bentuk Umum metode Runge-Kutta orde n

𝑦 = 𝑦 + 𝑎 𝑘 + 𝑎 𝑘 + ⋯ + +𝑎 𝑘

dimana 𝑎 , 𝑎 , ⋯ , 𝑎 adalah konstanta dan

𝑘 = ℎ𝑓 𝑡 , 𝑦
𝑘 = ℎ𝑓 𝑡 + 𝑝 ℎ, 𝑦 + 𝑞
𝑘 = ℎ𝑓 𝑡 + 𝑝 ℎ, 𝑦 + 𝑞 𝑘 + 𝑞 𝑘
⋮
𝑘 = ℎ𝑓 𝑡 + 𝑝 ℎ, 𝑦 + 𝑞( , ) 𝑘 + 𝑞( , ) 𝑘 + ⋯ + 𝑞( , )𝑘

Nilai-nilai 𝑎 , 𝑝 , 𝑞 ditentukan sedemikian hingga kesalahan tiap langkah

integrasi dapat diminimumkan.
4

2
Second Order Runge-Kutta Method
K1  h f ( xi , yi )
K 2  h f ( xi   h, yi   K1 )
yi 1  yi  w1K1  w2 K 2
Problem :
Find  ,  , w1 , w2
such that yi 1 is as accurate as possible.

Taylor Series in Two

Variables

3
Taylor Series in One Variable
The n th order Taylor Series expansion of f(x)


n
h i (i ) h n 1
f ( x  h)  f ( x)  f ( n 1) ( x )
i! ( n  1)!
i 0

Approximation Error

where x is between x and x  h

7

Derivation of 2nd Order

Runge-Kutta Methods – 1 of 5

Second Order Taylor Series Expansion

dy
Used to solve ODE :  f ( x, y )
dx
dy h 2 d 2 y
yi 1  yi  h  2
 O(h3 )
dx 2 dx
which is written as :
h2
yi 1  yi  h f ( xi , yi )  f ' ( xi , yi )  O ( h 3 )
2
8

4
Derivation of 2nd Order
Runge-Kutta Methods – 2 of 5

where f ' ( x, y ) is obtained by chain - rule differentiation

f ( x, y ) f ( x, y ) dy f f
f ' ( x, y )     f ( x, y )
x y dx x y
Substituting :
 f f  h2
yi 1  yi  f ( xi , yi )h    f ( xi , yi )   O ( h 3 )
 x y  2

Taylor Series in Two Variables

 f f 
f ( x  h, y  k )  f ( x, y )   h  k  
 x y 
1  2 2 f 2
2  f 2 f 
h  k  2 hk   ...
2!  x 2 y 2 x y 


n
i n 1
1    1    
  h  k  f ( x, y )   h  k  f ( x, y)
i!  x y  ( n  1)!  x y 
i 0
approximation error

( x , y ) is on the line joining between ( x, y ) and ( x  h, y  k )

10

5
Derivation of 2nd Order
Runge-Kutta Methods – 3 of 5

Problem : Find  ,  , w1, w2 such that

K1  h f ( xi , yi )
K 2  h f ( xi   h, yi   K1 )
yi 1  yi  w1K1  w2 K 2

Substituting :
yi 1  yi  w1h f ( xi , yi )  w2h f ( xi   h, yi   K1 )

11

Runge-Kutta Order 2 Method

f (t i  h, x i  (hf (t i , x i )))
xi   (hf (ti , xi ) )
x
f (ti , xi ) xi 1  xi  fi h
xi
fi fi  1hf (ti , xi ) 
2 hf (ti  h, xi  hf (ti , xi ))
h Weighted average of two slopes

x
ti ti+αh ti+1

6
Derivation of 2nd Order
Runge-Kutta Methods – 4 of 5

f f
f ( xi  h, yi   K1 )  f ( xi , yi )  h   K1  ...
x y
Substituting :
 f f 
yi 1  yi  w1h f ( xi , yi )  w2h  f ( xi , yi )  h  K1  ... 
 x y 
 f f 
yi 1  yi  ( w1  w2 )h f ( xi , yi )  w2h  h  K1  ... 
 x y 
f f
yi 1  yi  ( w1  w2 )h f ( xi , yi )  w2h 2  w2 h 2 f ( xi , yi )  ...
x y

13

Derivation of 2nd Order

Runge-Kutta Methods – 5 of 5

We derived two expansions for yi 1 :

f f
yi 1  yi  ( w1  w2 )h f ( xi , yi )  w2h 2  w2 h 2 f ( xi , yi )  ...
x y
 f f  h2
yi 1  yi  f ( xi , yi )h    f ( xi , yi )   O ( h 3 )
 x y  2
Matching terms, we obtain the following three equations :
1 1
w1  w2  1 , w2  , and w2  
2 2
3 equations with 4 unknowns  infinite solutions
1
One possible solution :     1, w1  w2 
2
14

7
2nd Order Runge-Kutta Methods
K1  h f ( xi , yi )
K 2  h f ( xi   h, yi   K1 )
yi 1  yi  w1K1  w2 K 2

Choose  ,  , w1 , w2 such that :

1 1
w1  w2  1, w2  , and w2  
2 2
15

Alternative Form
Second Order Runge Kutta
K1  h f ( xi , yi )
K 2  h f ( xi   h, y i   K1 )
yi 1  yi  w1K1  w2 K 2

Alternative Form
k1  f ( xi , yi )
k2  f ( xi   h, yi   h k1 )
yi 1  yi  h w1 k1  w2 k2  16

8
Choosing , , w1 and w2
1
For example, choosing   1, then   1, w1  w2 
2
Second Order Runge - Kutta method becomes :
K1  h f ( xi , yi )
K 2  h f ( xi  h, yi  K1 )
1
yi 1  yi 
2

K1  K 2   yi  h f ( xi , yi )  f ( xi 1, yi01 )
2

This is Heun ' s Method with a Single Corrector

17

Choosing , , w1 and w2
1 1
Choosing   then   , w1  0, w2  1
2 2
Second Order Runge - Kutta method becomes :
K1  h f ( xi , yi )
h K
K 2  h f ( xi  , yi  1 )
2 2
h K
yi 1  yi  K 2  yi  h f ( xi  , yi  1 )
2 2
This is the Midpoint Method
18

9
 2nd Order Runge-Kutta Methods
Alternative Formulas
1 1
 w2  ,  w2  , w1  w2  1
2 2
1 1
Pick any nonzero  number :    , w2  , w1  1 
2 2

Second Order Runge Kutta Formulas (select  0)

K1  h f ( xi , yi )
K 2  h f ( xi   h, yi   K1 )
 1  1
yi 1  yi  1   K1  K2
 2  2
19

K1  h f ( xi , yi )
Runge-Kutta Methods K 2  h f ( xi   h, y i   K1 )
1 1
w1  w2  1 , w2  , and w2  
2 2 yi 1  yi  w1K1  w2 K 2

Three of the most commonly used methods are:

yi 1  yi  w1K1  w2 K 2 ; K1  h f ( xi , yi )
 Huen Method with a Single Corrector (w2=1/2)
1 1
yi 1  yi  K1  K 2 ; K 2  h f ( xi  h, yi  K1 )
2 2
 The Midpoint Method (w2= 1)
1 1
yi 1  yi  K 2 ; K 2  h f ( xi  h, yi  K1 )
2 2
 Raltson’s Method (w2= 2/3)
1 2 3 3
yi 1  yi  K1  K 2 ; K 2  h f ( xi  h, yi  K1 )
3 3 4 4

10
Second order Runge-Kutta Method
Example
Solve the following system to find x (1.02) using RK2
x (t )  1  x 2  t 3 , x (1)  4, h  0.01,   1

STEP 1 :
K1  h f (t0  1, x0  4)  0.01(1  x02  t03 )  0.18
K 2  h f (t0  h, x0  K1 )
 0.01(1  ( x0  0.18)2  (t0  .01)3 )  0.1662
x (1  0.01)  x (1)  K1  K 2  / 2
 4  (0.18  0.1662) / 2  3.8269
21

Second order Runge-Kutta Method

Example

STEP 2

2 3
K1  h f (t1  1.01, x1  3.8269)  0.01(1  x1  t1 )  0.1668
K 2  h f (t1  h, x1  K1 )
 0.01(1  ( x1  0.1668)2  (t1  .01)3 )  0.1546
1
x (1.01  0.01)  x (1.01)  K1  K 2 
2
1
 3.8269  (0.1668  0.1546)  3.6662
2

22

11
 x (t )  1  x 2 (t )  t 3 , x (1)  4,
Solution for t  [1,2]
Using RK2,   1

CISE301_Topic8L4&5 23

Applications of Runge-Kutta Methods

to Solve First Order ODEs

Using Runge-Kutta methods of different

orders to solve first order ODEs

24

12
2nd Order Runge-Kutta RK2

Typical value of   1, Know as RK2

Equivalent to Heun' s method with a single corrector
k1  f ( xi , yi )
k2  f ( xi  h, yi  k1 h )
h
yi 1  yi  k1  k2 
2
Local error is O ( h 3 ) and global error is O ( h 2 )

25

Higher-Order Runge-Kutta

Higher order Runge-Kutta methods are available.

Derived similar to second-order Runge-Kutta.

Higher order methods are more accurate but

require more calculations.

26

13
3rd Order Runge-Kutta RK3

Know as RK3
k1  f ( xi , yi )
h 1
k2  f ( xi  , yi  k1h )
2 2
k3  f ( xi  h, yi  k1h  2k2h )
h
yi 1  yi  k1  4k2  k3 
6
Local error is O (h 4 ) and Global error is O ( h 3 )
27

4th Order Runge-Kutta RK4

k1  f ( xi , yi )
h 1
k2  f ( xi  , yi  k1h )
2 2
h 1
k3  f ( xi  , yi  k2h )
2 2
k4  f ( xi  h, yi  k3h )
h
yi 1  yi  k1  2k2  2k3  k4 
6
Local error is O ( h5 ) and global error is O (h 4 )
28

14
Higher-Order Runge-Kutta
k1  f ( xi , yi )
1 1
k2  f ( xi  h, yi  k1h )
4 4
1 1 1
k3  f ( xi  h, yi  k1h  k2 h )
4 8 8
1 1
k4  f ( xi  h, yi  k2 h  k3h )
2 2
3 3 9
k5  f ( xi  h, yi  k1h  k4h )
4 16 16
3 2 12 12 8
k6  f ( xi  h, yi  k1h  k2h  k3h  k4h  k5h )
7 7 7 7 7
h
yi 1  yi  7k1  32k3  12k4  32k5  7k6 
90
29

Example
4th-Order Runge-Kutta Method RK4

dy
 1 y  x2
dx
y ( 0 )  0 .5
h  0.2
Use RK 4 to compute y (0.2) and y (0.4)

30

15
Example: RK4
Problem :
dy
 1 y  x2 , y (0)  0.5
dx
Use RK 4 to find y (0.2), y (0.4)

31

4th Order Runge-Kutta RK4

k1  f ( xi , yi )
h 1
k2  f ( xi  , yi  k1h )
2 2
h 1
k3  f ( xi  , yi  k2h )
2 2
k4  f ( xi  h, yi  k3h )
h
yi 1  yi k1  2k2  2k3  k4 
6
Local error is O ( h5 ) and global error is O (h 4 )
32

16
Example: RK4 See RK4 Formula
Problem : h  0.2
dy
 1 y  x2 , y (0)  0.5 f ( x, y )  1  y  x 2
dx
x0  0, y0  0.5
Use RK 4 to find y (0.2), y (0.4)
k1  f ( x0 , y0 )  (1  y0  x0 2 )  1.5
1 1
k2  f ( x0  h, y0  k1h )  1   y0  0.15   x0  0.1  1.64
2
2 2
Step 1

1 1
k3  f ( x0  h, y0  k2h )  1   y0  0.164    x0  0.12  1.654
2 2
k4  f ( x0  h, y0  k3h)  1   y0  0.16545   x0  0.2 2  1.7908
h
y1  y0  k1  2k2  2k3  k4   0.8293
6
33

Example: RK4
Problem :
h  0.2
dy
 1 y  x2 , y (0)  0.5 f ( x, y )  1  y  x 2
dx
x1  0.2, y1  0.8293
Use RK 4 to find y (0.2), y (0.4)
k1  f ( x1, y1 )  1.7893
1 1
k2  f ( x1  h, y1  k1h )  1.9182
2 2
Step 2

1 1
k3  f ( x1  h, y1  k2 h )  1.9311
2 2
k4  f ( x1  h, y1  k3h )  2.0555
0.2
y2  y1  k1  2k2  2k3  k4   1.2141
6
34

17
Example: RK4
Problem :
dy
 1  y  x2 , y (0)  0.5
dx
Use RK 4 to find y (0.2), y (0.4)

Summary of the solution

xi yi
0.0 0.5
0.2 0.8293
0.4 1.2141
35

Summary
 Runge Kutta methods generate an
accurate solution without the need to
calculate high order derivatives.
 Second order RK have local truncation
error of order O(h3) and global truncation
error of order O(h2).
 Higher order RK have better local and
global truncation errors.
 N function evaluations are needed in the
Nth order RK method.
36

18