Skip to Main content Skip to Navigation
Journal articles

Régulation par apprentissage : application au conditionnement d'air

Abstract : The aim of this article is to demonstrate methodology making if possible to use data provided by a sensor network via machine learning methods in order to control building services automatically in a smart manner taking into account the physical properties of the building as well actual use and occupant comfort. Firstly, the methodology and experimental approach are described. Then the inital results of the implementation of a predictive control law within a test building at Berkeley University are presented. A significant decrease in the energy consumption required for air conditionning was mesured, and extrapolation to the entire building demonstrates true potential efficiency.
Document type :
Journal articles
Complete list of metadata
Contributor : Gilles Guerassimoff Connect in order to contact the contributor
Submitted on : Monday, June 11, 2018 - 5:57:29 PM
Last modification on : Wednesday, November 17, 2021 - 12:31:05 PM


  • HAL Id : hal-01812756, version 1


Gilles Guerassimoff, Ghassene Jebali, Therese Peffer. Régulation par apprentissage : application au conditionnement d'air. Revue Générale du Froid et du Conditionnement d'Air, 2018, ISSN 1958-4490 (1168), pp.22-31. ⟨hal-01812756⟩



Record views