pymgrid

pymgrid (PYthon MicroGRID) is a python library to generate and simulate a large number of microgrids.

Installation

Use the package manager pip to install pymgrid. You can clone and cd in the repo and then do:

pip install .

You can also run this pip command:

pip install git+https://github.com/Total-RD/pymgrid/

And in Google Colab:

!pip install git+https://github.com/Total-RD/pymgrid/

Getting Started

You can easily import the library from pip, and then import MicrogridGenerator from pymgrid.

from pymgrid import MicrogridGenerator as mg

m_gen=mg.MicrogridGenerator()
m_gen.generate_microgrid()

By default, this command will let you generate 10 microgrids. The object m_gen will have a list of microgrids that you can use.

First, you can get the control information with this command:

m_gen.microgrids[0].get_control_info()

pymgrid contains OpenAI Gym environments, you can use the following command to generate one:

from pymgrid.Environments.pymgrid_cspla import MicroGridEnv
from pymgrid import MicrogridGenerator as m_gen

#these line will create a list of microgrid
env = m_gen.MicrogridGenerator(nb_microgrid=25)
pymgrid25 = env.load('pymgrid25')
mg = pymgrid25.microgrids

#you can pass any of the microgrid to environment class:
env = MicroGridEnv({'microgrid':mg[0]})

#example of codes to to interact with the environment:
episode_reward = 0
done = False
obs = env.reset()
while not done:
    action = #your algorithm to select the next action
    obs, reward, done, info = env.step(action)
    episode_reward += reward

The control_dict dictionnary it the main way you will interact with the microgrid class. It