Contribution at the Smart Grid Comm 2021
Niklas Ebell presented his paper on “Benchmarking a Decentralized Reinforcement Learning Control Strategy for an Energy Community” at the hybrid IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids on Oct 27th 2021 at the E.On Energy Research Center in Aachen.
The aim of this work is to compare new approaches of multi-agent reinforcement learning with a rule-based and optimization based method in order to have a methodology at hand that allows to classify machine learning methods with respect to their performance. The application is a small energy community consisting of ten prosumers that pursue a common goal of a balanced energy cell (according to the cellular concept).