The introduction of building information modeling (BIM) in the design and monitoring of public buildings has seen increasing use over time. Likewise, there has been a growing interest in the IoT paradigm within the construction industry. This study aims to integrate the two approaches to optimize the electricity consumption of a public building through the interpretation of data collected by dedicated sensors. The article proposes the study of a process applied to a real case study and discusses and comments on the preliminary results obtained. Finally, a meeting point is found between the IoT technology and the BIM methodology through the definition of digital twin (DT), understood as a digital copy of practical reality both in the design phase and in monitoring and forecasting. The type of data collected by IoT sensors commonly falls under the big data paradigm, which is generally not analyzable through traditional techniques. Therefore, in order to extract new knowledge from historical data, deep learning techniques have been used, which are able to analyze and identify relationships between data in an intuitive way that otherwise could not be detected.
IoT and Deep Learning for Smart Energy Management
Cecere L.;Colace F.;Lorusso A.;Marongiu F.;Pellegrino M.;Santaniello D.
2024-01-01
Abstract
The introduction of building information modeling (BIM) in the design and monitoring of public buildings has seen increasing use over time. Likewise, there has been a growing interest in the IoT paradigm within the construction industry. This study aims to integrate the two approaches to optimize the electricity consumption of a public building through the interpretation of data collected by dedicated sensors. The article proposes the study of a process applied to a real case study and discusses and comments on the preliminary results obtained. Finally, a meeting point is found between the IoT technology and the BIM methodology through the definition of digital twin (DT), understood as a digital copy of practical reality both in the design phase and in monitoring and forecasting. The type of data collected by IoT sensors commonly falls under the big data paradigm, which is generally not analyzable through traditional techniques. Therefore, in order to extract new knowledge from historical data, deep learning techniques have been used, which are able to analyze and identify relationships between data in an intuitive way that otherwise could not be detected.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.