N-body solar system written in Python.

Andrew Hamill 39047415

Imports:

matplotlib - plotting
numpy - array maths
os - used to run terminal commands
platform - used to determine operating system
pickle - binary file storage

spiceypy - transformation matrices
astrpoy - gets data from specified ephemeris - source: https://github.com/poliastro/poliastro/archive/main.zip
poliastro - gets mass constants for planetary bodies

If using a environment manager, packages such numpy and matplot can be installed via CLI or GUI,

1. verify the correct environment is used,

conda info --envs
conda activate env-name

2. list and install packages,

conda list
conda isntall pkg-name

Other packages can be obtained through pip, Python's package manager:
pip install jplephem spiceypy
Poliastro must be obtained from Github,
pip install https://github.com/poliastro/poliastro/archive/main.zip

File structure:

Constants.py and Particle.py

• These handle mass constants, and the particle class used for all planetary bodies

Setup.py

• The main computation file, this handles class creation, initialising variables as well as user input for simulation parameters
  -> prompted when running Simulation.py as it contains a while loop
• Also contains Kepler's 3rd law calculations as this occurs after the simulation has run

Simulation.py

• The main simulation loop, graphing and outputs. This is the file you run.
• contains code and data saving used for graphs in the report, this is commented out as it's computationally expensive

Known issues:

• "Even when not re-running the simulation, I still have to enter simulation parameters?" - This was an oversight in implementation order causing the same while True: loop to run despite no intent to run the sim.
• Graphs included in sim file. - due to variable storage, they cannot be in their own file, as this would be dependent on Setup.py and Simulation.py, which creates circular dependencies.
• Kepler's law calculated outside of class - due to using post-sim. data, there's definitely a better way of doing this...
• .pkl cache file can become quite large, when doing long simulations (~ 1 GB)
• storing data for re-graphing can be very slow on low powered machnines - recommend commenting that out!

How to setup:

1. install necessary packages 1b) if using a lower powered computer (16GB ram recommended, 8GB min) comment out the code for saving data to a .pkl -> # save_pickle(data_store) on line 172 in Simulation.py
2. run Simulation.py
3. wait for ephemeris to download if not ran before
4. follow on screen dialogue