Skip to content

This Python project's goal is to model the collision of multiple galaxies through the use of an optimized N-Body problem engine.

License

Notifications You must be signed in to change notification settings

EnguerranVidal/GalaxyCollision

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

48 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

GALAXY MERGER

wakatime

This Python program's goal is to simulate the collision of multiple galaxies in a 2D plane and create a GIF animation of their interaction. For this simulation, we will use a modified N-Body Engine we used for the Planetary-Orbits-Solar-System Python project back in 2020.

The Galaxies :

The program's galaxies are made of a center massive black hole (white dot). It also contains big amount of massless particles representing stars that orbit around the center (blue dots). The particles are placed along circles/rings which share the same speed from Newtonian physics. We then give the galaxies an initial position and speed. An animation of the different speed rings can be found right above before the collision from the second galaxy..

The Particles :

As we mentioned earlier, the particles are separated into two groups.

  • Central Black Holes :

They are attractors and are also slowed down by galaxies' surrounding halos through dynamic friction, more can be found on Thijs Verkade's thesis "Simulating Galaxy Collisions in Python for Astronomy Education ". However, we will try to explain some of its principles down below. Galaxies are usually surrounded by a "halo" of dark matter and old stars which are distributed in a spherical cloud. When a massive body penetrates it, some dark matter and stars are put behind the massive body and slow it down.

The resulting acceleration is similar to air friction. It is usually called "dynamical friction" and is stronger the highest the massive body speed is. The used equation for this acceleration can be found right below, it direclty affects the massive body's velocity and ease the "merging" process for galaxies since without this force, colliding galaxies would just flung out to infinity from energy conservation. The density of the halo cloud is calculated using the method from the source thesis.

  • Massless Particles :

They represent stars which masses are insignificant when compared to the center black holes we discussed earlier. To ease the calculations the N-Body problem poses, we make them massless, which means they are not taken into account in the overall gravity calculations and do not affect the central black holes.

The N-Body Engine :

The calculations engine has been directly taken from the Planetary-Orbits-Solar-System Python Numerical Physics project I did with Jonathan OERS back in 2020. It was of course modified to fit the needs of this project, especially to separate the calculations of next-steps' positions and velocities for central massive bodies (black holes) and massless particles. The use of integration schemes to solve the N-Body problem differential equations is implemented through the Euler method and the Runge-Kutta 4 method which are the two most famous. We will try to implement more integratiosn schemes later and also try and compare those to the use of "odeint" from the scipy scientific library.

The Main Simulation Class :

The main class commands and runs calculation done by the above engine. It then stores the resulting data in a txt file put in the \logs directory with a header meant to portray the simulation parameters for easy replicability and further calculations through a system of "loadable sessions". Such header can be seen right below, it hosts some info about the diferent galaxies.

The data can then displayed in a Matplotlib animation. The animation can be stored too as a GIF file through the use of the imageio library. However, as of now, the GIF is not compressed and can therefore be quite heavy, we will try to fix this annoyance in the next update if possible.

CONTENTS

This repository contains the following files :

  • main.py : contains an runnable code resulting in the creation of the uncompressed top page animation of two galaxies colliding.
  • __galaxycollision.py : The main file containing the classes and functions used throughout the project.

INSTALLATION

git clone https://github.com/EnguerranVidal/Galaxy-Collision.git
cd Galaxy-Collision
pip install -r requirements.txt

USING THE CODE

To start a simulation run, the following command can be entered :

python3 main.py

To change variable such as galaxy size, number of rings, ass of the central massive bodies, incoming trajectories, feel free to change the code lines in main.py.

POSSIBLE UPDATES ?

  • The galaxies cannot, as of now, really "merg", this is due to the absence of a particle collision merg feature that could able us to fuse the multiple massive bodies when they get too close to one another. Since the main simulation class depends on the number of galaxies remaining constant, this feature could take a certain amount of time to implement. For now, the massive bodies simply get close and the least heavy is flung out of the system.

  • The created GIF from the Galaxy_Collision.display() method is for now uncompressed and can therefore become quite heavy, we will try to implement a compressing function later on to ease this problem.

  • Ttrails to the animation display will be added if possible. These will allow us to observe the course of the colliding galaxy.

  • The user will be allowed to showcase the time passing through a display in the collision animation as well as enter the wanted figure size.

  • The issue of units for masses and distances will be fixed since as of now, they are unspecified.

  • The simulation runs in a 2D plane which does not reflect the beauty of galaxy collisions. We will try to implement a 3D version later on.

Releases

No releases published

Packages

No packages published

Languages