A Review of Multi-agent Based Energy Management Systems

Gazafroudi, Amin Shokri; Paz, Juan F. De; Prieto‐Castrillo, Francisco; Villarrubia, Gabriel; Talari, Saber; Shafie‐khah, Miadreza; Catalão, João P.S.

doi:10.1007/978-3-319-61118-1_25

Cited by 16 publications

(5 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With the motivation to avoid a priori modelling of the underlying system dynamics (as required in a model-predictive controller), a model-free multi-agent reinforcement learning approach was given [10] -where [11] presented a general review of multi-agent energy management systems and [12] of multi-agent reinforcement learning. [13][14][15] also proposed a model-free reinforcement learning, namely Q-learning and deterministic policy gradient, multi-energy management system.…”

Section: Discussionmentioning

confidence: 99%

Model-predictive control and reinforcement learning in multi-energy system case studies

et al. 2021

View full text Add to dashboard Cite

Section: Discussionmentioning

confidence: 99%

Model-predictive control and reinforcement learning in multi-energy system case studies

et al. 2021

View full text Add to dashboard Cite

“…With the motivation to avoid a priori modelling on the underlying system dynamics (as required in a model-predictive controller), a model-free multi-agent reinforcement learning approach was given [10] where [11] presented a general review of multi-agent energy management systems and [12] of multi-agent reinforcement learning. [13][14][15] also proposed a model-free reinforcement learning, namely Q-learning and deterministic policy gradient, multi-energy management system.…”

Section: Discussionmentioning

confidence: 99%

Model-predictive control and reinforcement learning in multi-energy system case studies

Ceusters¹,

Rodríguez²,

García³

et al. 2021

Preprint

View full text Add to dashboard Cite

Model-predictive-control (MPC) offers an optimal control technique to establish and ensure that the total operation cost of multi-energy systems remains at a minimum while fulfilling all system constraints. However, this method presumes an adequate model of the underlying system dynamics, which is prone to modelling errors and is not necessarily adaptive. This has an associated initial and ongoing project-specific engineering cost. In this paper, we present an on-and off-policy multi-objective reinforcement learning (RL) approach, that does not assume a model a priori, benchmarking this against a linear MPC (LMPC -to reflect current practice, though non-linear MPC performs better) -both derived from the general optimal control problem, highlighting their differences and similarities. In a simple multi-energy system (MES) configuration case study, we show that a twin delayed deep deterministic policy gradient (TD3) RL agent offers potential to match and outperform the perfect foresight LMPC benchmark (101.5%). This while the realistic LMPC, i.e. imperfect predictions, only achieves 98%. While in a more complex MES system configuration, the RL agent's performance is generally lower (94.6%), yet still better than the realistic LMPC (88.9%). In both case studies, the RL agents outperformed the realistic LMPC after a training period of 2 years using quarterly interactions with the environment. We conclude that reinforcement learning is a viable optimal control technique for multi-energy systems given adequate constraint handling and pre-training, to avoid unsafe interactions and long training periods, as is proposed in fundamental future work.

show abstract

“…In this work, we present the multi agent-based structure for BEMS. There are several works that implement agents in the BEMS but none of does not consider a manner to consider residents' comfort in the BEMS [3][4][5][6][7][8]. Readers can refer to to know more about multi-agent systems.…”

Section: Proposed Schemementioning

confidence: 99%

RETRACTED CHAPTER: An Agent-Based Model for Energy Management of Smart Home: Residences’ Satisfaction Approach

Ebrahimi

Hajizade

2019

Advances in Intelligent Systems and Computing

View full text Add to dashboard Cite

Reducing the cost of energy leads to reduction in resident's comfort. Therefore, it is necessary to consider both reduction in cost and resident's dissatisfaction. Smart buildings include different systems such as communication system (CS), sensing system (SS), grid data collecting system (GDCS), building energy management system (BEMS), hybrid system (HS), temporary service systems (TSS) and permanent service systems (PSS). Considering the aforementioned systems, an agent-based model for energy management of smart buildings is presented in this work to reduce the energy cost as well as the resident's dissatisfaction.

show abstract

A Review of Multi-agent Based Energy Management Systems

Cited by 16 publications

References 19 publications

Model-predictive control and reinforcement learning in multi-energy system case studies

Model-predictive control and reinforcement learning in multi-energy system case studies

Model-predictive control and reinforcement learning in multi-energy system case studies

RETRACTED CHAPTER: An Agent-Based Model for Energy Management of Smart Home: Residences’ Satisfaction Approach

Contact Info

Product

Resources

About