1

Part 5: Integration and

Differentiation

Chapter 19: Numerical

Integration

ENG-307
2 Objectives
 Recognizing that Newton-Cotes integration formulas are based on
the strategy of replacing a complicated function or tabulated data
with a polynomial that is easy to integrate.
 Knowing how to implement the following single application
Newton-Cotes formulas:
 Trapezoidal rule
 Simpson’s 1/3 rule
 Simpson’s 3/8 rule

 Knowing how to implement the following composite Newton-Cotes

formulas:
 Trapezoidal rule
 Simpson’s 3/8 rule

 Knowing how to use the trapezoidal rule to integrate unequally

spaced data.
 Understanding the difference between open and closed integration
formulas.
3 Introduction
Integration:

 f x dx
b
I a

is the total value, or summation, of f(x) dx over the range from

a to b:
4 Newton-Cotes Formulas

 The Newton-Cotes formulae are some of the most common numerical

integration schemes.
 They are based on the strategy of replacing a complicated function or
tabulated data 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

 Consider for example,

the approximation of an
integral by the area under:
 A single straight line 
 A single parabola  
5 Newton-Cotes Formulas
 Or… the approximation of an
integral by the area under
three straight lines 

 Closed and open forms of the

Newton-Cotes formulae are
available:
a. Closed forms: data at the
beginning and end of the
limits of integration are
known.
b. Open forms: Integration
limits extend beyond range
of data.
6 The Trapezoidal Rule

The trapezoidal rule is the

first of the Newton-Cotes
closed integration formulas;
it uses a straight-line
approximation for the
function:
b b
I   f ( x)dx   f1 ( x)dx
a a

 f b   f a  
x  a  dx
b
I    f (a) 
a
 ba 
f a   f b 
I  b  a 
2
7 The Trapezoidal Rule

 An estimate for the local

truncation error of a single
application of the trapezoidal
rule is:
1
Et   f  b  a
3

12
where is somewhere between a
and b.
 This formula indicates that the
error is dependent upon the
curvature of the actual function
as well as the distance between
the points.
 Error can thus be reduced by
breaking the curve into parts.
8 The Trapezoidal Rule

 The local truncation error (true error) in the Trapezoidal rule

(Derivation in text):

 Since, , (unknown exactly), take an average value of

the
 The approximate error is:

Where:

 Note: 
9 The Trapezoidal Rule

Example:
Use trapezoidal rule to estimate the integral of:

f (x) = 0.2 + 25x − 200x2 + 675x3 − 900x4 + 400x5

from a = 0 to b = 0.8. Estimate the error. Recall that the exact value of the integral is
1.640533.
1
0
The Trapezoidal Rule

 Assuming n+1 data points are evenly

spaced, there will be n intervals over
which to integrate.
 The total integral can be calculated by
integrating each subinterval and then
adding them together:
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
xn x1 x2 xn
I  f n  x  dx   f n  x  dx   f n  x  dx     f  x  dx
x0 x0 x1 xn 1 n

f  x0   f  x1  f  x1   f  x2  f  xn 1   f  xn 
I   x1  x0    x2  x1      xn  xn 1 
2 2 2
h n 1 
I   f  x0   2  f  xi   f  xn  
2 i 1 
1
1
The Trapezoidal Rule

An error for multiple-application “composite” trapezoidal rule can be obtained

by summing the individual errors for each segment:
(b  a)3
Ea   3  f ( i)
12n
The average of the 2nd order summation can be obtained by integration over
the interval, divided by interval, and multiplied by number intervals, or
b
 f ( i)  nf  ( f    f ( )d / (b  a ) )
a

(b  a )3
Ea   2
f 
12n

So if the number of segments is doubled, the truncation error will be

quartered!
12 The Trapezoidal Rule

Example:
Use the two-segment (composite) trapezoidal rule to estimate the integral of:

f (x) = 0.2 + 25x − 200x2 + 675x3 − 900x4 + 400x5

from a = 0 to b = 0.8. Estimate the error. Recall that the exact value of the integral is
1.640533.
13 The Trapezoidal Rule – MATLAB

h ba

2 2n
14 The Trapezoidal Rule – MATLAB
 Simpson’s Rules

 One drawback of the trapezoidal rule is that the error is

related to the second derivative of the function.
 More complicated approximation formulas can improve the
accuracy for curves - these include using (a) 2nd and (b) 3rd
order polynomials.
 The formulas that result from taking the integrals under
these polynomials are called Simpson’s rules.
16 Simpson’s 1/3 Rule
 In this method, the function is approximated by a 2nd
order polynomial (quadratic- parabola) .

 Requires : 3 points connected by a parabola (odd)

& 2 segments (even)
f2 (x)

f(x)

x0 x1 x2 x
h h
17 Simpson’s 1/3 Rule
 Using 2nd order Lagrange polynomial:

Let , & , → # of segments

 After integration and algebraic manipulation:

 Simpson’s 1/3 Rule
The most accurate formula for equally spaced 2 segments.
18 Simpson’s 1/3 Rule
 The true error is given by (see text for derivation):

 As with the trapezoidal rule, since we do not know the value of ,

we will use instead of :

 The approximate error is:

 Note: 
19 Simpson’s 1/3 Rule
• Example
Integrate the function

from 0 to 0.8 using Simpson’s 1/3 rule with 2 segments.

Solution: ( 1.640533)
b  a 0.8  0
2 segments → =2 h   0.4
n 2
8
f ( 0)  0. 2 f (0.4)  2.456 f (0.8)  0.232
h
I   f (0)  4 f (0.4)  f (0.8)   03.4 0.2  4(2.456)  0.232  1.367467
3
1.640533  1.367467
t   16.64%
1.640533
20 Simpson’s 1/3 Rule
 Multiple-applications of Simpson’s 1/3 rule (Text 21.2):
a. divide the interval to smaller segments.

b. Apply the method for each segment

AAAA AAAA AAAA

AAAA AAAA AAAA

AA AA AA AA AAAA
21 Simpson’s 1/3 Rule
 Multiple-applications of Simpson’s 1/3 rule (Text 21.2):
c. For uniform spacing (equal segments):

AA A AA A
, ,.. , ,.. AAA AA A

AA A AAAA AAAA AA A
, ,.. , ,..

 The true error of the multiple-application Simpson’s 1/3 rule is:

 The approximate error is:

Where:
22 Simpson’s 1/3 Rule
 Note:
a. We need an even number of segments, , for multiple
applications of Simpson’s 1/3 rule because we need 2
segments (i.e. 3 points) for each application of the rule.
it is limited to situations where there are an even number of
segments and odd number of points
b. Simpson’s rule is superior to trapezoidal rule for most
applications.
c. However, it is limited to cases where values are equal spaced
data.
23 Simpson’s 1/3 Rule
← 2 segments.

Multiple
application
, ,.. , ,..
24 Simpson’s 1/3 Rule
• Example
Integrate the function

from 0 to 0.8 using Simpson’s 1/3 rule with 4 segments.

Solution: ( 1.640533)
b  a 0 .8  0
4 segments → =4 h   0 .2
n 4
f (0)  0.2 f (0.2)  1.288 f (0.4)  2.456
f (0.6)  3.464 f (0.8)  0.232
h
I   f ( 0 )  4 ( f ( 0 .2 )  f ( 0 . 6 ))  2 f ( 0 . 4 )  f ( 0 . 8 ) 
3
0.2
  0.2  4(1.288  3.464)  2(2.456)  0.232  1.623467
3
1.640533 1.623467
t   1.04%
1.640533
25 Simpson’s 3/8 Rule
 In this method, our function is approximated by a 3rd order
(cubic) polynomial.
→ we need 4 points (even) 3 segments (odd).
→ will use points a and b , in addition to two
more points in between them

 Approximating by the 3rd order polynomial then integrating:

Using 3rd order Lagrange polynomial:
26 Simpson’s 3/8 Rule

Substitute for and integrate

AA
AA

← Simpson’s
3/8 Rule

The most accurate formula for equally spaced 3 segments.

27 Simpson’s 3/8 Rule
 The true error of the multiple-application Simpson’s 3/8 rule is:

 As with the trapezoidal and Simpson’s 1/3 rules, since we do not

know the value of , we will use instead of :

 The approximate error is:

 Note: 
28 Simpson’s 3/8 Rule
• Example
Integrate the function

from 0 to 0.8 using the most accurate method for 3 segments.

Solution: ( 1.640533)
3 segments  = 3  the most accurate methods is Simpson’s 3/8 rule
b  a 0.8  0
h   0.2667
n 3
f (0)  0.2 f (0.2667)  1.432774 f (0.5334)  3.487177 f (0.8)  0.232
3h
 I   f (0)  3( f (0.2667)  f (0.5334))  f (0.8)
8
3 * 0.2667
 0.2  3(1.43277  3.487177)  0.232  1.51917
8
1.640533  1.51917
t   7.4%
1.640533
29 Simpson’s 3/8 Rule
 Note: 𝐸 , / =− 𝑓̅ 𝑥 𝐸 , / =− 𝑓̅ 𝑥

a) The error for the Simpson’s 3/8 rule is same as that for the
Simpson’s 1/3 rule except that we have a larger denominator in the
Simpson’s 3/8 rule (6,480 compared to 2,880).  The Simpson’s
3/8 rule is somewhat more accurate than the Simpson’s 1/3 rule.

b) Even though the Simpson’s 3/8 rule is somewhat more accurate

than the Simpson’s 1/3 rule, it still has the same order of accuracy
& at the same time, it requires 4 points compared to 3 points only
for the Simpson’s 1/3.  The Simpson’s 1/3 rule is usually
preferred over the Simpson’s 3/8 rule.

c) When the # of segments is odd, the Simpson’s 3/8 rule can be

mixed with the Simpson’s 1/3 rule. In this case, use Simpson’s 3/8
rule to the last 3 segments and Simpson’s 1/3 rule is applied to all
the previous segments.
30 Simpson’s 3/8 Rule
• Example
Determine the integral of the following tabular data using Simpson’s
rules: 0 0.1 0.2 0.3 0.4 0.5
1 7 4 3 5 9

Solution: We have five segments →

Use the 1/3 rule for the 1st two segments.

Use the 3/8 rule for the last 3 segments.

Rule Rule
31 Simpson’s 3/8 Rule
32 Quick Review
1. The trapezoidal rule: Linear
a. 2 points: 1 segment

b. Multiple-applications:

2. Simpson’s 1/3 Rule: 2nd order

a. 3 points 2 segments:

b. Multiple-applications:
, ,.. , ,..

3. Simpson’s 3/8 Rule: cubic

4 points 3 segments:
33 Integration with Unequal Segments
 All the previous discussion is for closed integrals with
equally spaced data.
 Sometimes experimental data are not equally spaced.
 For non-equally spaced data, we apply Trapezoidal
rule to each segment and then sum the results.
   
xn x1 x2 xn
I fn x  dx  fn x  dx  fn x  dx  fn x  dx
x0 x0 x1 x n1

f x0   f x1  f x1   f x2  f xn1  f xn 

I  x1  x0   x2  x1   xn  xn1 
 2 2 2
34 Integration with Unequal Segments
 Integral of Unevenly-Spaced Data:
a. Trapezoidal rule for each segment or:
b. Could also be evaluated with Simpson’s rule for
higher accuracy.
35 Integration with Unequal Segments
• Example x f(x)
0.0 0.200000
Integrate the following data:
0.12 1.309729
Apply Trapezoidal rule: 0.22 1.305241
0.32 1.743393
0.36 2.074903
0.40 2.456000
h1 h2
I   f(x0 )  f(x1 )   f(x1 )  f(x2 ) 0.44 2.842985
2 2 0.54 3.507297
hn
    f(xn 1 )  f(xn )
0.64 3.181929
2 0.70 2.363000
0.80 0.232000

1.309729  0.2 1.305241  1.309729 0.232  2.363

0.12  0.10    0.10
2 2 2
36 Integration with Unequal Segments
• Example
Integrate the following data. Use the most accurate methods.
x f(x)
The most accurate formula
0.0 0.200000
TR 1 segment 0.12
0.12 1.309729

1/3 rule 2 segments 0.10 0.22 1.305241

0.32 1.743393
0.36 2.074903
3/8 rule 3 segments 0.04
0.40 2.456000
0.44 2.842985
1/3 rule 2 segments 0.10 0.54 3.507297
0.64 3.181929
TR 1 segments 0.06 0.70 2.363000
TR 1 segments 0.10 0.80 0.232000
Integration Code for Unequal Segments
MATLAB Functions
 MATLAB has built-in functions to evaluate
integrals based on the trapezoidal rule
 z = trapz(y)
 z = trapz(x, y)
produces the integral of y with respect to x. If x
is omitted, the program assumes h=1.
 z = cumtrapz(y)
 z = cumtrapz(x, y)
produces the cumulative integral of y with
respect to x. If x is omitted, the program
assumes h=1.
39 Integration with Unequal Segments
40 Integration with Unequal Segments

Suppose that we had measurements of velocity at a series of discrete unequally spaced

times during free fall (70-kg jumper with 0.275 kg/m drag coefficient). A nice application
of integration is to compute the distance z(t) of an object based on its velocity v(t).

Analytical