A state-space model to control an adaptive facade prototype using data-driven techniques - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Energy and Buildings Année : 2023

A state-space model to control an adaptive facade prototype using data-driven techniques

Résumé

In this article, a framework to model and predict the energy performance of an adaptive facade is proposed. A case study of a bio-inspired concept, called Stegos, is considered. This dynamic system manages thermal transfers through the facade by varying the color and position of rotating flaps. A prototype of this concept that was incorporated into a test bench was tested at a 1:1 scale and in real weather conditions, while the flaps color and angle were changed manually. The objective of the article was twofold. First, using measurements, a reduced order model was identified by applying the Modal Identification Method (MIM). The training phase was divided into four consecutive steps. At each step, one day of corresponding experimental data is used. The reduced model provided reliable predictions of heat flux values induced by the prototype when the flaps were in a closed or fully open state. Second, a Model Predictive Control (MPC) was implemented to indicate the optimal configurations of the prototype for better energy efficiency. Case study used measurements of one week in winter and determined the color and angle of the flaps, which corresponded to the optimal solution. Closed black flaps during a day and open flaps during a night contributed to maximum heat gain.
Fichier principal
Vignette du fichier
Stegos_article_EnB.pdf (5.55 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04273114 , version 1 (07-11-2023)

Identifiants

Citer

Ainagul Jumabekova, Julien Berger, Tessa Hubert, Antoine Dugué, Tingting Vogt Wu, et al.. A state-space model to control an adaptive facade prototype using data-driven techniques. Energy and Buildings, 2023, 296, pp.113391. ⟨10.1016/j.enbuild.2023.113391⟩. ⟨hal-04273114⟩
38 Consultations
43 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More