Title :

Energy Management for Microgrids: a Reinforcement Learning Approach

Author(s) :

Levent, Tanguy [Auteur]
Laboratoire de physique des interfaces et des couches minces [Palaiseau] [LPICM]
Preux, Philippe [Auteur]
Sequential Learning [SEQUEL]
Le Pennec, Erwan [Auteur]
Modélisation en pharmacologie de population [XPOP]
Centre de Mathématiques Appliquées - Ecole Polytechnique [CMAP]
Badosa, Jordi [Auteur]
Laboratoire de Météorologie Dynamique (UMR 8539) [LMD]
Henri, Gonzague [Auteur]
TOTAL S.A.
Bonnassieux, Yvan [Auteur]
Laboratoire de physique des interfaces et des couches minces [Palaiseau] [LPICM]

Conference title :

ISGT-Europe 2019 - IEEE PES Innovative Smart Grid Technologies Europe

City :

Bucharest

Country :

France

Start date of the conference :

2019-09-29

Journal title :

Proc. ISGT-Europe 2019 - IEEE PES Innovative Smart Grid Technologies Europe

Publisher :

IEEE

English keyword(s) :

Index Terms-Microgrid
Energy Management System
Agent Based
Supervised Learning
Reinforcement Learning

HAL domain(s) :

Informatique [cs]/Intelligence artificielle [cs.AI]
Statistiques [stat]/Machine Learning [stat.ML]
Sciences de l'ingénieur [physics]/Energie électrique

English abstract : [en]

This paper presents a framework based on reinforcement learning for energy management and economic dispatch of an islanded microgrid without any forecasting module. The architecture of the algorithm is divided in two parts: ...
Show more >This paper presents a framework based on reinforcement learning for energy management and economic dispatch of an islanded microgrid without any forecasting module. The architecture of the algorithm is divided in two parts: a learning phase trained by a reinforcement learning (RL) algorithm on a small dataset and the testing phase based on a decision tree induced from the trained RL. An advantage of this approach is to create an autonomous agent, able to react in real-time, considering only the past. This framework was tested on real data acquired at Ecole Polytechnique in France over a long period of time, with a large diversity in the type of days considered. It showed near optimal, efficient and stable results in each situation.Show less >

Language :

Anglais

Peer reviewed article :

Oui

Audience :

Internationale

Popular science :

Non

Collections :

• Centre de Recherche en Informatique, Signal et Automatique de Lille (CRIStAL) - UMR 9189

Source :

Harvested from HAL