Etude de stratégies de gestion énergétique des bâtiments par l'application de la programmation dynamique

Abstract : The subject of this work is the control of building equipments with the use of a dynamic programming optimization. This algorithm allows to develop a predictive controller in order to reduce the energy counsumption of the building or to increase the comfort of the occupants.A first case study is developed with the control of the heating system in order to shift the load from peak hours to off-peak hours according to utility rate incentives or carbon emissions.A second case study is about the control of stores and mechanical ventilation in order to increse comfort in the building during a strong heat wave summer week. The control of openings, like windows and doors, is also developed to use natural ventilation in order to increase comfort. The results of such a control strategy are compared with another control strategy set up in a test house "Air et Lumière".The last part of this work is about weather forecast of local climatic data by using Markov chains and artificial neural networks. The influence of weather forecast errors on the developed control strategy is also studied.
Document type :
Theses
Complete list of metadatas

https://pastel.archives-ouvertes.fr/pastel-00957327
Contributor : Abes Star <>
Submitted on : Monday, March 10, 2014 - 11:02:10 AM
Last modification on : Monday, November 12, 2018 - 10:59:56 AM
Long-term archiving on : Tuesday, June 10, 2014 - 11:45:40 AM

File

2013ENMP0039.pdf
Version validated by the jury (STAR)

Identifiers

  • HAL Id : pastel-00957327, version 1

Collections

Citation

Bérenger Favre. Etude de stratégies de gestion énergétique des bâtiments par l'application de la programmation dynamique. Autre. Ecole Nationale Supérieure des Mines de Paris, 2013. Français. ⟨NNT : 2013ENMP0039⟩. ⟨pastel-00957327⟩

Share

Metrics

Record views

1619

Files downloads

6383