Part 6

Numerical Integration and

Differentiation

1
Mathematics - Integral

2
 What is Integration?

• Integrate means “to bring together”, as parts, into a whole;

to indicate total amount.

b
I   f ( x ).dx
a

• The above stands for integral of function f(x) with respect

to the independent variable x between the limits x = a to x
= b.

3
 Chapter 21

Newton-Cotes Integration
Formulas

4
 What is Integration?

• Graphically integration is simply to find the area under a

certain curve between the 2 integration limits.

b
I   f ( x ).dx  A
a

5
 Newton-Cotes Integration Formulas
Introduction
• The Newton-Cotes formulas are the most
common numerical integration methods.
• They are based on the strategy of replacing a
complicated function with an approximating
function that is easy to integrate.
b b
I   f ( x)dx   f n ( x)dx
a a

f n ( x)  a0  a1 x    an 1 x n 1  an x n

6
 1. Trapezoidal Rule
The trapezoidal rule uses a polynomial of the first
degree to replace the function to be integrated.
b b
I   f ( x ).dx   f
a a
1 ( x ).dx

f (b)  f ( a )
f1 ( x )  a  ( x  a)
b b
ba
I   f ( x ).dx   f 1 ( x ).dx
a a
b
 f (b)  f ( a ) 
  a  ( x  a ) .dx
a 
ba 
f (a)  f (b)
I  (b  a)
2
7
 Error of the Trapezoidal Rule

When we employ the integral under a straight

line segment to approximate the integral under a
curve, error may be:

1


Et   f ( )(b  a) 3

12

Where x lies somewhere in the interval from a to

b.

8
Trapezoidal Rule
Example 21.1

9
 Multiple Trapezoidal Rule

• One way to improve the accuracy of the trapezoidal rule

is to divide the integration interval from a to b into a
number of segments and apply the method to each
segment.

• The areas of individual segments can then be added to

yield the integral for the entire interval.

10
Multiple Trapezoidal Rule

11
 Multiple Trapezoidal Rule
ba
h a  x0 b  xn
n
x1 x2 xn

I  f ( x)dx   f ( x)dx     f ( x)dx

x0 x1 xn1

Substitute into the integrals for f(x) by f1(x) in

each segment and integrate:
f ( x0 )  f ( x1 ) f ( x1 )  f ( x2 ) f ( xn 1 )  f ( xn )
I h h  h
2 2 2
h n 1

I   f ( x0 )  2 f ( xi )  f ( xn )
2 i 1 
12
13
 Multiple Trapezoidal Rule

An error for multiple-application trapezoidal rule can be

obtained by summing the individual errors for each
segment:

 f (i)  nf 
(b  a)3
Ea   2
f 
12n

14
 Simpson’s Rules

More accurate estimate of an integral is obtained if a high-

order polynomial is used to connect the points. The
formulas that result from taking the integrals under such
polynomials are called Simpson’s Rules.

15
 Simpson’s Rules

• Simpson’s 1/3 Rule

Results when a second-order interpolating polynomial is
used.

• Simpson’s 3/8 Rule

Results when a third-order (cubic) interpolating polynomial
is used.

16
 Lagrange Interpolating Polynomials

• The general form for n+1 data points is:

n
f n ( x)   Li ( x) f ( xi )
i 0
n x  xj
Li ( x)  
j 0 xi  x j
j i

designates the “product of”

17
 Lagrange Interpolating Polynomials

• Linear version (n = 1):

Used for 2 points of data: (xo,f(xo)) and (x1,f(x1)),

x  x1 x  x0
f1 ( x)  f ( x0 )  f ( x1 )
x0  x1 x1  x0

Lo (x) L1 ( x)

18
 Lagrange Interpolating Polynomials

• Second order version (n = 2):

f 2 ( x) 
 x  x1  x  x2 
f ( x0 ) Lo ( x) , j  0
x0  x1 x0  x 2 

 x  x0  x  x2 
f ( x1 ) L1 ( x) , j  1
x1  x0 x1  x 2 

 x  x0  x  x1 
f ( x2 ) L2 ( x) , j  2
x2  x0 x2  x1 
19
 Simpson’s Rules

Simpson’s 1/3 Rule Simpson’s 3/8 Rule

20
 Simpson’s 1/3 Rule

b b
I   f ( x)dx   f 2 ( x)dx
a a

a  x0 b  x2
x2
 ( x  x1 )( x  x2 ) ( x  x0 )( x  x2 ) ( x  x0 )( x  x1 ) 
I   f ( x0 )  f ( x1 )  f ( x2 )dx
x0 
( x0  x1 )( x0  x2 ) ( x1  x0 )( x1  x2 ) ( x2  x0 )( x2  x1 ) 

h ba
I   f ( x0 )  4 f ( x1 )  f ( x2 ) h
3 2

Simpson’s 1/3 Rule

21
 Simpson’s 1/3 Rule

• Single segment application of Simpson’s 1/3 rule has

a truncation error of:

(b  a) 5 ( 4)
Et   f ( ) a  b
2880
• Simpson’s 1/3 rule is more accurate than trapezoidal
rule.

22
The Multiple-Application Simpson’s 1/3 Rule
• Just as the trapezoidal rule, Simpson’s rule
can be improved by dividing the integration
interval into a number of segments of equal
width.
f ( xo )  4 f ( x1 )  f ( x2 ) f ( x 2 )  4 f ( x 3 )  f ( x4 )
I  2h  2h
6 6
f ( x n  2 )  4 f ( x n 1 )  f ( x n ) ba
..........  2h with h 
6 n
 n 1 n2 
 f ( xo )  f ( xn )  4  f ( xi )  2  f ( x j ) 
 (b  a )  
i  1, 3 , 5 j  2 , 4 ,6

3n
(b  a) 5 ( 4) Error is fourth derivative means the
Et   f ( )
180n 4 method is third order accurate
23
The Multiple-Application Simpson’s 1/3 Rule
Graphical representation of
the multiple application of
Simpson’s 1/3 rule. Note that
the method can be employed
only if the number of segments
is even.

24
 Simpson’s 3/8 Rule
If there are 2 extra points between the integration
limits a and b, then a 3rd degree polynomial can
be used instead of the parabola to replace the
function to be integrated.

25
 Simpson’s 3/8 Rule

If there are 2 extra points between the integration

limits a and b, then a 3rd degree polynomial can
be used instead of the parabola to replace the
function to be integrated:
b b
I   f ( x)dx   f 3 ( x)dx
a a

3h (b  a)
I   f ( x0 )  3 f ( x1 )  3 f ( x2 )  f ( x3 ), h 
8 3
(b  a) 5 ( 4)
Et   f ( ) Simpson’s 3/8 Rule
6480
26
The Simpson’s 3/8
rule is used in
conjunction with the
1/3 rule to permit
evaluation
of both even and odd
numbers of segments.

27
 Newton Cotes Integration-Example
Find the integral of:
f(x) = 0.2 +25 x – 200 x 2 + 675 x 3 – 900 x 4 + 400 x 5
Between the limits 0 to 0.8, f(0) = 0.2, f(0.8) = 0.232,
Iexact=1.640533
1. The trapezoidal rule (ans. 0.1728)
f (a)  f (b) 0.2  0.232
I  (b  a)  I  (0.8  0)  0.1728
2 2
Et  1.640533  0.1728  1.467733   t  89.5%
f '' ( x)  400  4050 x  10,800 x 2  8000 x 3
0 .8

f ''
( x) 

0
(400  4050 x  10,800 x 2  8000 x 3 ) dx
 60
0.8  0
1
Ea   (60)(0.8)3  2.56
12 28
 Newton Cotes Integration-Example

2. Multiple trapezoidal rule (n=4) (ans. 1.4848)

f(0)=0.2, f(0.2)=1.288, f(0.4)=2.456, f(0.6)=3.464,
f(0.8)=0.232
(b  a) (0.8  0)
h   0.2
4 4

h n 1

I   f ( x0 )  2 f ( xi )  f ( xn )
2 i 1 
0.2
 0.2  2(1.288  2.456  3.464)  0.232  1.4848
2

29
 Newton Cotes Integration-Example
3. The Simpson 1/3 rule (ans. 1.367467)
f(0) = 0.2, f(0.4) = 0.2.456, f(0.8) = 0.232
b  a 0.8  0
h   0.4
2 2
h
I   f ( x0 )  4 f ( x1 )  f ( x2 )
3
0.4
 0.2  4  2.456  0.232  1.367467
3
Et  1.640533 1.367467  0.2730667   t  16.6%
f ( 4) ( x)  2400

(b  a)5 ( 4) (0.8  0)5

Ea   f ( )   (2400)  0.2730667
2880 2880
30
 Newton Cotes Integration-Example
4. Multiple application of Simpson 1/3 rule (n=4)
(ans. 1.623467).
f(0)=0.2, f(0.2)=1.288, f(0.4)=2.456, f(0.6)=3.464 ,f(0.8)=0.232
(b  a) (0.8  0)
h   0.2
4 4
h n 1 n2

I   f ( x0 )  4  f ( xi )  2  f ( xi )  f ( xn )
3 i 1, 3, 5 i  2 , 4.6 
0.2
 0.2  4(1.288  3.464)  2(2.456)  0.232  1.623467
3
Et  1.640533 1.623467  0.017067   t  1.04%
(b  a ) 5 ( 4 ) 0.85
Ea   4
f ( )   4
(2400)  0.017067
180n 180(4)
31
 Newton Cotes Integration-Example

5. The Simpson 3/8 rule (ans. 1.519170)

f(0)=0.2, f(0.2667)=1.432724, f(0.5333)=3.487177, f(0.8)=0.232
(b  a) (0.8  0)
h   0.2667
3 3
3h
I  I   f ( x0 )  3 f ( x1 )  3 f ( x2 )  f ( x3 )
8
0.8
 0.2  3 1.432724  3  3.487177  0.232  1.519170
8
Et  1.640533 1.51917  0.121363   t  7.4%

(b  a)5 ( 4) 0.85
Ea   f ( )   (2400)  0.1213630
6480 6480

32
 Chapter 22

Integration of Equations

33
 Romberg Method

• Trapezoid formula with a sub-interval h gives an error

of the order O(h2).

• We can combine two Trapezoid estimates with

intervals h and h/2 to get a better estimate.

34
 Romberg Method

R(0,0)
ba
R (0,0)   f ( a )  f (b) R(1,0) R(1,1)
2 R(2,0) R(2,1) R(2,2)
ba
h n
, R(3,0) R(3,1) R(3,2) R(3,3)
2
2( n 1) 
1 
R( n,0)  R( n  1,0)  h 
2  
 k 1

f a  ( 2k  1)h 



R ( n, m )  m
1
4 1
 
4m  R( n, m  1)  R( n  1, m  1) n  1, m  1

38
 Property of Romberg Method

Theorem R(0,0)
b R(1,0) R(1,1)


a
f ( x)dx  R(n, m)  O(h 2 m 2 ) R(2,0) R(2,1) R(2,2)
R(3,0) R(3,1) R(3,2) R(3,3)

Error Level O(h2 ) O(h4 ) O(h6 ) O(h8 )

39
 Example

1
Compute
 x 2dx
0
0.5
3/8 1/3

ba
h  1, R (0,0)   f (a )  f (b)  1 0  1  0.5
2 2
1 1 11 11 3
h  , R (1,0)  R (0,0)  h ( f ( a  h ))       
2 2 22 24 8

R ( n, m )  m
1
4 1
 
4m  R ( n, m  1)  R ( n  1, m  1) for n  1, m  1

1 1  3 1 1
R (1,1)  1 4  R (1,0)  R (0,0)  4    
4 1 3  8 2 3

40
 Example (cont.)
0.5

3/8 1/3

11/32 1/3 1/3

1 1 1  3 1  1 9  11
h , R ( 2,0)  R (1,0)  h( f ( a  h )  f ( a  3h ))        
4 2 2  8  4  16 16  32

R ( n, m )  m
1
4 1
 
4m  R( n, m  1)  R ( n  1, m  1)

1 1  11 3  1
R( 2,1)  4  R ( 2,0)  R (1,0)  4    
3 3  32 8  3

R( 2,2)  2
1
4 1
 1 16 1  1
42  R( 2,1)  R (1,1)     
15  3 3  3

41
 When do we stop?

STOP if
R ( n, n )  R ( n, n  1)  
or
After a given number of steps,
for example, STOP at R(4,4)

42
 Gauss Quadrature

• Gauss quadrature implements a strategy of positioning

any two points on a curve to define a straight line that
would balance the positive and negative errors.
• Hence, the area evaluated under this straight line
provides an improved estimate of the integral.
 Two points Gauss-Legendre
Formula
• Assume that the two Integration points are xo and x1 such that:
ba
2
I  f ( x ).dx  c
ba
1 f ( x1 )  c2 f ( x2 )

2

• The object of Gauss quadrature is to determine the equations

of the form:

I  c0 f ( x0 )  c1 f ( x1 )
• c0 and c1 are constants, the function arguments x0 and x1 are
unknowns…….(4 unknowns)
 Two points Gauss-Legendre
Formula

• Thus, four unknowns 1

c0 f ( x0 )  c1 f ( x1 )   1 dx  2
to be evaluated require 1
four conditions. 1
c0 f ( x0 )  c1 f ( x1 )   x dx  0
• If this integration is 1
1
2
exact for a constant, c0 f ( x0 )  c1 f ( x1 )   x dx 
2

3
1st order, 2nd order, and 1
1
3 order functions:
rd
c0 f ( x0 )  c1 f ( x1 )   x 3 dx  0
1
 Two points Gauss-Legendre
Formula
• Solving these 4 equations, we can determine c1,
c2, x1 and x2.
The weighting factors are : c0  c1  1
The Integratio n points are :
1
x0    0.5773503
3
1
x1   0.5773503
3
 1   1 
I  f  f  
 3  3
 Two points Gauss-Legendre
Formula
• Since we used limits for the previous integration from –1
to 1 and the actual limits are usually from a to b, then we
need first to transform both the function and the
integration from the x-system to the xd-system
f()
x b 1 f(x)
f(x1)

x  a    1 f(xo)
ba ba
x  
2 2
ba x
dx  d xo x1
2 a b 
-1 1
 Higher-Points Gauss-Legendre
Formula
1
I   f ( )d
1
i using n Gauss points :

n
I   ci f ( i )  c1 f ( 1 )  c2 f ( 2 )  .....  cn f ( n )
i 1
 Multiple Points Gauss-Legendre
Points Weighting factor Function argument Exact for
2 1.0 -0.577350269 up to 3rd
1.0 0.577350269 degree

3 0.5555556 -0.774596669 up to 5th

0.8888889 0.0 degree
0.5555556 0.774596669

4 0.3478548 -0.861136312 up to 7th

0.6521452 -0.339981044 degree
0.6521452 0.339981044
0.3478548 0.861136312

6 0.1713245 -0.932469514 up to 11th

0.3607616 -0.661209386 degree
0.4679139 -0.238619186
0.4679139 0.238619186
0.3607616 0.661209386
0.1713245 0.932469514
 Gauss Quadrature - Example

Find the integral of:

f(x) = 0.2 + 25 x – 200 x2 + 675 x3 – 900 x4 + 400 x5
Between the limits 0 to 0.8 using:

– 2 points integration points (ans. 1.822578)

– 3 points integration points (ans. 1.640533)