Lab Manual

Course Code: MCC538

Couse Name: Computer Oriented Numerical Methods

Objective: To provide knowledge for implementation of Numerical Methods in Python programming.

Experiments

Exp.No.1: Basics of Python programming, Implementation of read/write files, Python Functions

and Module.
Aim: To write functions and modules with inbuilt libraries.
Numerical producers:
Step-1: Use Mathematical Functions, reading input, printing output, opening and closing of files, reading
writing data to and from file.
Step-2: Write functions and modules.
Output: As per the use functions/modules/read or write file.
Example-1: To make an program to create data file and again read that data file and plot/print data.
Example-2: To make a .txt file of 100 random numbers between two points.

Exp.No.2: Python program for Newton’s forward formula and central Difference Interpolation
Formulae.
Aim: Find the value of a function at an intermediate point.
Numerical producers:
Step-1: The first stage of cubic spline interpolation is to set up equation and solve them for the unknown k0 s.
This task is carried out by the function curvatures.
Step-2: The second stage is the computation of the interpolation at x. This step can be repeated any number
of times with different values of x using the function evalSpline.
Output: As per the program we get intermediate point value.

Example-1: Find the value of sin(52) using Newton’s forward formula. The data table is given below:

θ 45◦ 50◦ 55◦ 60◦

sin(θ ) 0.7071 0.7660 0.8192 0.8660

Example-2: Find the value of f (28) using central difference interpolation formula. The data table is gi-
ven below:

θ 20 25 30 35 40
f (θ ) 49225 48316 47236 45926 44306

1
 Exp.No.3: Python program for Adaptive Integration.
Aim: How to take adaptive step size(h) in an integration method.
Numerical producers:
Step-1: Set the tolerance for integration method.
Step-2: If error is grater then tolerance then take step size(h) is half of step size(h).
Step-3: If error is less then tolerance then take step size(h) is double of step size(h).
Output: It is a part of integration method which use for time efficiency and less computation efforts.
Example: Find the solution of given system of ODE using any of the IVP integration method with adaptive
step size:

ẋ = σ (y − x)
ẏ = x(ρ − z) − y
ż = xy − β z
[x0 , y0 , z0 ] = [1, 1, 1]
[σ , ρ, β ] = [10, 28, 8/3]
time = 0 to 40

Exp.No.4: Python program for Gaussian Quadrature.

Aim: Find integration of simple algebraic function using Gaussian Quadrature.
Numerical producers:
n
Step-1: I = ∑ Ai f (xi ) is known as Newton-Cotes rules. The difference lies in the way that the weights Ai
i=0
and nodal abscissas xi are determined.
Step-2: In Gaussian quadrature the nodes and weights are chosen so that above equation yields the exact
integral if f (x) is a polynomial of degree 2n + 1 or less.
Output: We get the integration of given simple algebraic function
5
Example: Evaluate 3 − 11(1 − x2 ) 2 dx as accurately as possible with Gaussian integration.

Exp.No.5: Python program for Multiple Integrals and Solution of Integral Equations.
Aim: Find multiple integral of a function.
Numerical producers:
Step-1: Take a required function for integration.
Step-2: Write a program for multiple integration or using python libraries.
Output: Solution of required function after multiple integration.
Example: Find the value of (x3 + y2 + xy)dx dy.
RR

2
 Exp.No.6: Python program for Numerical Solution of linear Equations.
Aim: To solve n linear, algebraic equations in n unknowns using Python.
Numerical producers:
Step-1: Take n linear, algebraic equations in n unknowns.
Step-2: Select Direct or Indirect Method.
Step-3: Using function, and required libraries write program.
Output: Solution of required function after multiple integration.
Example: Find the solution of x − 6y + z = 0; 3x + 2z = 0; 2x + 9y = 0.

Exp.No.7: Python program for Solution of Non-linear Equations.

Aim: To solve required Non-linear Equations i.e. calculate the values of x and y so that f1 (x, y) = 0 and
f2 (x, y) = 0.
Numerical producers:
Step-1: Take a required Non-linear Equations.
Step-2: Using function and required libraries write program.
Step-3: Solve by proper algorithm.
Output: Output gives us the required solution of Non-linear Equations.
Example: Find the root of the given non-linear equation up to four decimal place accuracy by the method of
Newton-Raphson method.

f1 (x, y) = x + xy − 4
f2 (x, y) = x + y − 3

Exp.No.8: Python program Solution of Ordinary Differential Equations using Explicit Runge-
Kutta Methods.
Aim: Find the solution of Ordinary Differential Equations using Explicit Runge-Kutta Methods.
Numerical producers:
Step-1: Take a required Ordinary Differential Equation.
Step-2: Using function and required libraries (scipy) solve the problem.
Output: Output gives us the required solution of Ordinary Differential Equation.
Example-1: Integrate the ODE by fourth-order Runge-Kutta method

δy
= 3y − 4ex − x : y(0) = 1
δx
from x = 0 to 10 in steps of h = 0.1.

3
 Exp.No.9: Python program Solution of Ordinary Differential Equations using Explicit Runge-
Kutta with Adaptive Step Size.
Aim: Find the solution of Ordinary Differential Equations using Explicit Runge-Kutta Methods with Adap-
tive Step Size.
Numerical producers:
Step-1: Take a required Ordinary Differential Equation.
Step-2: Using function and required libraries (scipy) solve the problem.
Output: Output gives required solution of Ordinary Differential Equation with Adaptive Step Size.
Example: Integrate the given equation
√
ÿ + 2ẏ + 3y = 0; y(0) = 0; ẏ(0) = 2
√
with the adaptive Runge-Kutta method from x = 0 to 5 (the analytical solution is y = e sin 2x).

Exp. No.10: Python program for Multi-step methods, Extrapolation methods.

Aim: Solving polynomial or ODE Multi-step Methods, Extrapolation methods.
Numerical producers:
Step-1: Take a required polynomial.
Step-2: Using function and required libraries solve polynomial by Multi-step Methods and Extrapolation
methods.
Output: Output gives required solution of polynomial.
Example-1: Compute the solution of the initial value problem ẏ = sin y : y(0) = 1 at x = 0.5 with the midpoint
formulas using n = 2 and n = 4, followed by Richardson extrapolation.

Exp. No.11: Python program for Shooting Method, Boundary value problems.
Aim: Solving Boundary value problems.
Numerical producers:
Step-1: Take a Boundary value problems.
Step-2: Using function and required libraries solve Boundary value problem by Shooting Method.
Output: Output gives required solution of Boundary value problem.
Example: Solve differential equation by BVP Method

y000 + 2y00 + sin(y) = 0

y(−1) = 0; y0 (−1) = −1; y0 (1) = 1

4
 Exp. No.12: Python program for Numerical Solutions of Partial Differential Equations (PDEs).
Aim Take a Boundary value problems.
Numerical producers:
Step-1: Take a Boundary value problems.
Step-2: Using function and required libraries solve Partial Differential Equations
Output: Output gives required solution of Partial Differential Equations (PDEs).
Example: Solve Laplace equation,

∂ 2u ∂ 2u
+ =0
∂ 2x ∂ 2y
where, u(x, y) defined on x ∈ [0, 1], y ∈ [0, 1] with the boundary conditions
u(x, 0) = 1, u(x, 1) = 2, u(0, y) = 1, u(0, y) = 2.

Lab Exam.13: