/Electric_Vehicle_Charging_Simulation

This project implements Q-Learning to find the optimal policy for charging and discharging electric vehicles in a V2G scheme under conditions of uncertain commitment of EV owners. The problem is modelled as a multi-objective multi-agent cooperative game. Project is part of fulfillment criteria for ECE 730 course at the University of Alberta.

Primary LanguagePython

Electric Vehicle Charging Simulation

To run the simulation make sure the AESO_2020_demand_price.csv file is in the same directory as the simulation.py file.

See the report here.

To run use:

python simulation.py --n --id_run --pen --scale
  • n: number of iterations, default 10 (int)\
  • id_run: file name to save results, default 'test' (str)\
  • pen: the market penetration of EVs, in number of EVs, default 0.1 (float)\
  • scale: the scaling factor for the model, default 1000 (int)

stats_output_with_v2g.py and stats_output_no_v2g.py are used to generate statistics after the model has been trained.

Requrements:

  • numpy
  • pandas
  • tqdm
  • argprase