[en] The problem faced by the operator of a storage device participating in a continuous intra-day (CID) market is addressed in this paper. The goal of the storage device operator is the maximization of the cumulative rewards received over the entire trading horizon, while taking into account operational constraints. The energy trading is modeled as a Partially Observable Markov Decision Process. An equivalent state representation and high-level actions are proposed in order to tackle the variable number of the existing orders in the order book. The problem is solved using deep reinforcement learning (RL). Preliminary results indicate that the agent converges to a policy that scores higher total revenues than the ``rolling intrinsic''.
Disciplines :
Electrical & electronics engineering
Author, co-author :
Boukas, Ioannis ; Université de Liège - ULiège > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Smart-Microgrids
Ernst, Damien ; Université de Liège - ULiège > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Smart grids
Papavasiliou, Anthony; Université Catholique de Louvain - UCL > Center for Operations Research and Econometrics > CORE > Associate professor
Cornélusse, Bertrand ; Université de Liège - ULiège > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Smart-Microgrids
Language :
English
Title :
Intra-day Bidding Strategies for Storage Devices Using Deep Reinforcement Learning
Publication date :
June 2018
Event name :
15th International Conference on the European Energy Market
Event organizer :
Polish Association of Electrical Engineers Lodz University of Technology
Event place :
Łódź, Poland
Event date :
from 27-06-2018 to 29-06-2018
Audience :
International
Main work title :
International Conference on the European Energy Market, Łódź 27-29 June 2018