Python Course: DSC4-P

Dr. Puja Mondal, Department of Physics, Vivekananda College, Thakurpukur, Kolkata-63

Lecture 10: Numerical Integration: Trapezoidal and Simson's 1/3 method

Trapezoidal method
In this method, the whole range of integration is divided into an n
number of equally spaced subintervals of width h.

The value of the integral in the first subinterval is the area of the
trapezium with base h and bounded by two parallel lines y and y : 0 1

,
h(y0 +y1 )

where h
xn −x0
=
n

xn h
∫ ydx = [y0 + 2(y1 + y2 +. . . . . . . +yn−1 ) + yn ]
x0 2

Evaluate the following integrations using the trapezoidal

rule
i) ∫ ii) ∫
2.5 π/4
2 4 1/4
(1 + x + x ) dx √tan(x) dx
0 0

taking the number of intervals as runtime input.

For each case verify your answer using the corresponding function from
the SciPy library.

Also, plot the integrand functions as f(x) vs x in the range of the given
integration limit using matplotlib in separate figures.
In [1]: import numpy as np
from scipy.integrate import trapz
import matplotlib.pyplot as plt

def f(x):
return (1+x**2+x**4)**(1/4)
def f1(x):
return np.sqrt(np.tan(x))
n=int(input("No of intervals"))
print (n)
a,b=0, 2.5
a1, b1=0, np.pi/4
h=(b-a)/n
h1=(b1-a1)/n
I_1=0.5*h*(f(a)+f(b)+2*sum([f(a+i*h) for i in range(1,n)]))
I_2=0.5*h1*(f1(a1)+f1(b1)+2*sum([f1(a1+i*h1) for i in range(1,n)]))
print('Value of first integral=', I_1)
print('Value of second integral=', I_2)
# Compare the result with the result obtained using the Scipy function
x=np.linspace(a,b, num=n)
x1=np.linspace(a1,b1, num=n)
I1_tra=trapz(f(x),x)
I2_tra=trapz(f1(x1),x1)
print('Value of first integral using trapz function from Scipy library =', I1_tra)
print('Value of second integral using trapz function from Scipy library =', I2_tra)

# plot of the first integrand function

plt.plot(x,f(x),'--pg',label='f(x)=$(1+x^{2}+x^{4})^{1/4}$')
# linestyle=--, marker= 'p' (pentagon), color=green
plt.xlabel('x', size=20)
# set the size of x tick label font size
plt.xticks(size=15)
plt.ylabel('f(x)', size=20)
# set the size of y tick label font size
plt.yticks(size=15)
plt.legend(fontsize=14)
plt.show()

# plot of the second integrand function

plt.plot(x1,f1(x1),':*r',label='f1(x)=$\sqrt{tan(x)}$')
# linestyle=:, marker= '*', color=red
plt.xlabel('x1', size=20)
# set the size of x tick label font size
plt.xticks(size=15)
plt.ylabel('f1(x1)', size=20)
# set the size of y tick label font size
plt.yticks(size=15)
plt.legend(fontsize=14)
plt.show()

20
Value of first integral= 3.9961570250846172
Value of second integral= 0.4860062400095109
Value of first integral using trapz function from Scipy library = 3.9962907405062658
Value of second integral using trapz function from Scipy library = 0.4858907386385015
Simson's 1/3 method
In this method, the whole range of integration is divided into an n-even
number of equally spaced subintervals of width h.

In this method, we need two subintervals each of width h. The value of

the integration in the interval(-h,h) (consider the first two intervals) is
(ah + 6c) by considering the equation of parabola as
h 2

f (x) = ax
2
+ bx + c .
h
h h
2
∫ f (x) dx = (2ah + 6c) = [f (x0 ) + 4f (x1 ) + f (x2 ]
3 3
−h

where h = and we have three points (x in the two

xn −x0

n 0, x1 , x2 )

subintervals each of length h.

Summing over all subintervals, we get

xn
h
∫ f (x)dx = [f (x0 ) + 4(f (x1 ) + f (x3 )+. . . . +f (xn−1 )) + 2(f (x2 ) + f (x4 )+. . . . +f (xn−2 ))
3
x0

+ f (xn )]

Evaluate the following integrations using Simpson's 1/3 rule

i) ∫ ii) ∫
2π 5 2
2 −x /2
x sin(2x) dx e dx
0 −5

taking the number of intervals as runtime input.

For each case verify your answer using the corresponding function from
the SciPy library.
 Also, plot the integrand functions as f(x) vs x in the range of the given
integration limit using matplotlib in separate figures.
In [2]: n=20
print ([i for i in range(1, n,2)]) # odd values of n
print ([i for i in range(2, n-1, 2)]) # even values of n

[1, 3, 5, 7, 9, 11, 13, 15, 17, 19]

[2, 4, 6, 8, 10, 12, 14, 16, 18]

In [3]: import numpy as np

from scipy.integrate import simps
import matplotlib.pyplot as plt

def f(x):
return x**2*np.sin(2*x)
def f1(x):
return np.exp(-x**2/2)
n=int(input("No of intervals"))
print (n)
# Integration limit for the first integration
a,b=0, 2*np.pi
h=(b-a)/n
# First integration
f_odd=4*sum ([f(a+i*h) for i in range(1, n, 2)])
f_even=2*sum ([f(a+i*h) for i in range(2, n-1, 2)])
I1_simp=(h/3)*(f(a)+f(b)+f_odd+f_even)
print('Value of first integral=',I1_simp)
x=np.linspace(a,b, num=n)
I1_simp_scipy=simps(f(x),x)
print('Value of first integral using simps function from Scipy library =', I1_simp_scipy
# Second integration
a1,b1=-5.0, 5.0
h1=(b1-a1)/n
f1_odd=4*sum ([f1(a1+i*h1) for i in range(1, n, 2)])
f1_even=2*sum ([f1(a1+i*h1) for i in range(2, n-1, 2)])
I2_simp=(h1/3)*(f1(a1)+f1(b1)+f1_odd+f1_even)
print('Value of first integral=',I2_simp)
x1=np.linspace(a1,b1, num=n)
I2_simp_scipy=simps(f1(x1),x1)
print('Value of first integral using simps function from Scipy library =', I2_simp_scipy

# plot of the first integrand function

plt.plot(x,f(x),'--pg',label='f(x)=$x^{2}sin(2x)$')
# linestyle=--, marker= 'p' (pentagon), color=green
plt.xlabel('x', size=20)
# set the size of x tick label font size
plt.xticks(size=15)
plt.ylabel('f(x)', size=20)
# set the size of y tick label font size
plt.yticks(size=15)
plt.legend(fontsize=14)
plt.show()

# plot of the second integrand function

plt.plot(x1,f1(x1),':*r',label='f1(x)=$e^{-x^{2}/2}$')
# linestyle=:, marker= '*', color=red
plt.xlabel('x1', size=20)
# set the size of x tick label font size
plt.xticks(size=15)
plt.ylabel('f1(x1)', size=20)
# set the size of y tick label font size
plt.yticks(size=15)
 plt.legend(fontsize=14)
plt.show()

40
Value of first integral= -19.74028969475288
Value of first integral using simps function from Scipy library = -19.727365488719492
Value of first integral= 2.506626821938234
Value of first integral using simps function from Scipy library = 2.5066270343358035

In [ ]: