Reinforcement Learning for Stochastic Networks, Toulouse

Energy Management in Smart Grids Using Finite Horizon SOR Q-learning

Jun 19, 2024, 2:30 PM

30m

A001 (ENSEEIHT)

Parallel session: Some applications of reinforcement learning to networks

Speaker

Shalabh Bhatnagar (Indian Institute of Science)

Description

The smart grid is comprised of
different microgrids and is supported by different technologies. The microgrid has to make a lot of decisions, and this has to be automated for efficiency. We have Markov Decision Processes as a frame-
work for sequential decision-making under uncertainty and
reinforcement learning techniques to learn the optimal decisions. In this work, we formulate this problem in the setting of finite horizon
Markov Decision Processes and propose two
algorithms. One is a typical finite horizon
algorithm similar to [1], and the other is an improvement
of this algorithm using the technique of successive over
relaxation. This is then applied to the smart
grid problem.