Future Energy Systems: RL for energy optimization

Are you looking to reduce your electricity bill? Well you’re in luck. The massive growth of smart meters creates a new era in which intelligent systems can provide novel services to electricity customers. As people are becoming increasingly aware of their carbon footprint and rising electricity bill, many are looking to optimize their household energy use.

In this project, engineers in the Data-X Lab compared two deep reinforcement learning algorithms (DQN and DPG) in search for a better way to help homeowners reduce both their carbon footprint and their electricity costs. Using these algorithms, as well as a prediction method using neural networks (i.e. LSTM), the engineers were able to build the foundation of an energy management system, that could reduce your costs and emissions.

https://github.com/ashkanyousefi/Future_Energy_Systems