Deep reinforcement learning for home energy management system control

Lissa, Paulo; Deane, Conor; Schukat, Michael; Seri, Federico; Keane, Marcus; Barrett, Enda

doi:10.1016/j.egyai.2020.100043

Cited by 109 publications

(34 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For each successful reduction of consumption, the agent received a reward aimed at maximizing the amount of energy saved. Lissa et al (2021) developed such a model, based on Markov Decision Processes (MDP) to control the temperature of domestic hot water. Their goal was to reduce the consumption by optimizing the usage of energy produced by photo-voltaic panels.…”

Section: Deep Learningmentioning

confidence: 99%

Intelligent Energy Management Systems -- A Review

Mischos¹,

Eleanna²,

Vrakas³

2022

Preprint

View full text Add to dashboard Cite

Climate change has become a major problem for humanity in the last two decades. One of the reasons that caused it, is our daily energy waste. People consume electricity in order to use home/work appliances and devices and also reach certain levels of comfort while working or being at home. However, even though the environmental impact of this behavior is not immediately observed, it leads to increased CO2 emissions coming from energy generation from power plants. Confronting such a problem efficiently will affect both the environment and our society. Monitoring energy consumption in real-time, changing energy wastage behavior of occupants and using automations with incorporated energy savings scenarios, are ways to decrease global energy footprint. In this review, we study intelligent systems for energy management in residential, commercial and educational buildings, classifying them in two major categories depending on whether they provide direct or indirect control. The article also discusses what the strengths and weaknesses are, which optimization techniques do they use and finally, provide insights about how these systems can be improved in the future.

show abstract

Section: Deep Learningmentioning

confidence: 99%

Intelligent Energy Management Systems -- A Review

Mischos¹,

Eleanna²,

Vrakas³

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…The saving was greater when a neural network was trained for each zone of the building. Lissa et al [27] relied on that same neural network to estimate the Q-values of their deep reinforcement learning algorithm model to manage and control the heating system and the domestic hot water. They proposed a new methodology to define the threshold of indoor temperature comfort, based on the historical temperature of the building.…”

Section: B Reinforcement Learning and Its Growthmentioning

confidence: 99%

Designing a generalised reward for Building Energy Management Reinforcement Learning agents

Mulero

Arregi

Beitia

et al. 2021

2021 6th International Conference on Smart and Sustainable Technologies (SpliTech)

View full text Add to dashboard Cite

The reduction of the carbon footprint of buildings is a challenging task, partly due to the conflicting goals of maximising occupant comfort and minimising energy consumption. An intelligent management of Heating, Ventilation and Air Conditioning (HVAC) systems is creating a promising research line in which the creation of suitable algorithms could reduce energy consumption maintaining occupants' comfort. In this regard, Reinforcement Learning (RL) approaches are giving a good balance between data requirements and intelligent operations to control building systems. However, there is a gap concerning how to create a generalised reward signal that can train RL agents without delimiting the problem to a specific or controlled scenario. To tackle it, an analysis and discussion is presented about the necessary requirements for the creation of generalist rewards, with the objective of laying the foundations that allow the creation of generalist intelligent agents for building energy management.

show abstract

“…Several studies have shown that RL-based EMSs can be successfully implemented in various microgrid topologies, either as a single agent [19][20][21] or as a multiagent [22,23]. Nonetheless, the most basic and widely used RL approaches, namely Q-learning [24], suffer from several challenges, including inefficient data utilization, inability to handle continuous/large statespace, and curse of dimensionality, which cause the method to fail for large-scale tasks.…”

Section: Introductionmentioning

confidence: 99%

Deep Q‐network application for optimal energy management in a grid‐tied solar PV‐Battery microgrid

Muriithi

Chowdhury

2022

The Journal of Engineering

View full text Add to dashboard Cite

This paper presents a deep Q‐network (DQN) technique to optimally manage energy resources in a microgrid in which the algorithm learns tasks in the same way as humans do. Every move the agent makes in the environment generates feedback which then motivate the agent to learn more about the environment and perform far more intelligent steps later in its learning stages. This paper proposes a DQN‐based energy management system that learns system uncertainties, including load demand, grid prices and volatile renewable power supply to ensure that energy is optimally dispatched in such a setting. The method uses experience replay and target network to increase learning speed and improve stability in previous research. The performance of the system is evaluated with slow, medium, and fast fluctuating load profiles and their combinations and also with different sets of actions. The system is designed to minimize both grid transactional costs and battery degradation costs. Results shows that the proposed system successfully learns from experience to raise the battery state of charge and optimally shift loads from a one‐time instance, reducing aggregate peak load and cost of energy purchased from the utility grid while maximizing revenue from energy exported to the grid.

show abstract

Deep reinforcement learning for home energy management system control

Cited by 109 publications

References 16 publications

Intelligent Energy Management Systems -- A Review

Intelligent Energy Management Systems -- A Review

Designing a generalised reward for Building Energy Management Reinforcement Learning agents

Deep Q‐network application for optimal energy management in a grid‐tied solar PV‐Battery microgrid

Contact Info

Product

Resources

About