RL_HEMS

Reinforcement Learning for HEMS

This project documents the code used for our RL HEMS experiements using Q Learning and DQN.
In order to run these files successfully, it is necessary to do the following:

1. Download the zip file associated with this project or clone the repository
2. Have Python 3.9 installed on your machine
3. Run pip install -r requirements.txt to set up the environment for the files
4. Ensure that the following files are in the same directory as the .py files:
   • home.toml
   • 2019.npy
   • 2020.npy
   • 2021.npy
5. Run python HEMS_dqn.py to run the DQN algorithm for HEMS. This file completes 10 runs with explicitly set seeds to gather the results based on the data provided in Step 4. On an i7-1165G7 @ 2.80GHz machine, this operation takes around 3 hours to complete.
6. Run python HEMS_q_learning.py to run the Q Learning algorithm for HEMS. This file completes 10 runs with explicitly set seeds to gather the results based on the data provided in Step 4. On an i7-1165G7 @ 2.80GHz machine, this operation takes around 1.5 hours to complete.
7. Run python PPO_HEMS.py or python PPO_HEMSShiftableInt.py or PPO_HEMSShiftableUnInt.py to run the PPO algorithm for HEMS. This file completes only one run with explicitly set seeds. Warning: this code may take hours to complete on an i7-1165G7 @ 2.80GHz machine.