RK4_2
i. Description
ii. Dependencies
iii. Instructions
iv. Output
--------------------------------------------------------------------
i. DESCRIPTION:
The code in the current repository constitutes
a simple 2nd order ODE solver written in C with
additional plotting done with Python (NumPy + matplotlib).
The program essentially solves equations of the following form:
(d^2)y/(dx)^2 = f(t, x, v)
The numerical method used is Runge - Kutta 4th order, one of
the most well-known methods for the numerical solution of such
differential equations.
Newton's 2nd law formalism has been kept; the right-hand function
is of the form force(t, x, v)/mass.
--------------------------------------------------------------------
ii. DEPENDENCIES:
The program is written in C and is expected to compile and run
with no problems in any (linux) system.
Plotting is done numpy and matplotlib, two fundamental packages
for numerical Python.
Installing is straightforward for linux systems,
for example in Ubuntu:
sudo apt-get install python-numpy python-matplotlib
--------------------------------------------------------------------
iii. INSTRUCTIONS:
1. Enter the desired force function f(t, x, v) into the
F function, inside the src/RK4.c source file.
2. Enter the desired initial conditions for t, x and v at the beginning
of the program, in main (right after the memory allocations).
3. Compile, using the provided makefile. Simply type "make" in the
main directory, where the makefile is also located.
4. Run the program from the command line in the following format:
$ ./RK4_2 <timestep> <total time> <output interval>
For example, running the following:
$ ./RK4_2 0.1 100 10
will use a time step of 0.1, up to 100 time units and produce
output every 10 timesteps (i.e. every 10*0.1 = 1 time units).
---------------------------------------------------------------------
iv. OUTPUT:
The program outputs the independent variable t and the numerical
solutions x and v as columns in the file "output/out.txt".
It also plots the solution using matplotlib, producing the
following plots:
1. (t, x(t)) and (t, v(t)) as separate plots in a single figure.
2. (x(t), v(t)) (a.k.a. phase-space plot).
To do the above, enter the "analysis" directory and run:
$ python plot1D.py