A systematic literature review on the use of artificial intelligence in energy self-management in smart buildings
AuthorsAguilar Castro, José Lisandro; Garcés Jiménez, Alberto; Rodríguez Moreno, María Dolores; García, Rodrigo
IdentifiersPermanent link (URI): http://hdl.handle.net/10017/49380
Aguilar Castro, J., Garcés Jiménez, A., Rodríguez Moreno, M.D. & García, R., 2021, "A systematic literature review on the use of artificial intelligence in energy self-management in smart buildings", Renewable and Sustainable Energy Reviews, vol. 151, art. no. 111530.
Energy management system
Autonomous management architecture
Systematic literature review
info:eu-repo/grantAgreement/EC/H2020/754382/EU/GOT Energy Talent/GET
Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)
Buildings are one of the main consumers of energy in cities, which is why a lot of research has been generated around this problem. Especially, the buildings energy management systems must improve in the next years. Artificial intelligence techniques are playing and will play a fundamental role in these improvements. This work presents a systematic review of the literature on researches that have been done in recent years to improve energy management systems for smart building using artificial intelligence techniques. An originality of the work is that they are grouped according to the concept of "Autonomous Cycles of Data Analysis Tasks", which defines that an autonomous management system requires specialized tasks, such as monitoring, analysis, and decision-making tasks for reaching objectives in the environment, like improve the energy efficiency. This organization of the work allows us to establish not only the positioning of the researches, but also, the visualization of the current challenges and opportunities in each domain. We have identified that many types of researches are in the domain of decision-making (a large majority on optimization and control tasks), and defined potential projects related to the development of autonomous cycles of data analysis tasks, feature engineering, or multi-agent systems, among others.