Microgrid Trading Using Reinforcement Learning

This project originated as a graduation project and later evolved into three research papers presented at ICCCEEE20, available on IEEEXPLORE. It focuses on the management and control of a trading game between islanded microgrids using different deep reinforcement learning techniques. These techniques were implemented in a custom environment designed by our research team.

The papers present algorithms that serve as trading controllers for islanded microgrids, with data applied from Sudanese villages. We explored the use of two Deep Reinforcement Learning algorithms, DDPG and PPO, on the custom environment.

The code for this project can be found here.