Procédure d'identification pour la régulation prédictive basée sur un modèle appliquée à un immeuble de bureaux avec des systèmes pour bâtiment activés thermiquement.

Identification for model based predictive control applied to an office building with a thermally activated building systems.

Numéro : pap. 342

Auteurs : ZACEKOVA E., VANA Z., VERHELST C., et al.

Résumé

In recent years, the number of applications that use advanced control algorithms such as Model Predictive Control (MPC) for building climate control is increasing. These control strategies provide a substantial savings potential, but for their optimal behaviour it is necessary to have a good mathematical model of the building. Creation of a mathematical model appropriate for MPC is a delicate task. This is especially the case of real applications, characterized by a lack of available measurements and by poor quality data. This paper presents a system identification (SI) procedure to obtain a building model based on real measurement data. The innovative character of the presented SI method is mainly the choice of the identification procedure cost function - multi-step prediction error criterion - which directly corresponds to the MPC cost function. The resulting model is suitable for use within the MPC framework, where the model should be accurate even during long time horizons. Being used within the MPC, such a model outperforms classically identified models based on one-step prediction error minimization. The procedure is illustrated for the Hollandsch Huys building, a 3-storey office building, which is monitored and studied in the framework of the Geotabs project.

Documents disponibles

Format PDF

Pages : 10 p.

Disponible

  • Prix public

    20 €

  • Prix membre*

    15 €

* meilleur tarif applicable selon le type d'adhésion (voir le détail des avantages des adhésions individuelles et collectives)

Détails

  • Titre original : Identification for model based predictive control applied to an office building with a thermally activated building systems.
  • Identifiant de la fiche : 30008884
  • Langues : Anglais
  • Sujet : Environnement
  • Source : Clima 2013. 11th REHVA World Congress and 8th International Conference on Indoor Air Quality, Ventilation and Energy Conservation in Buildings.
  • Date d'édition : 16/06/2013

Liens


Voir d'autres communications du même compte rendu (424)
Voir le compte rendu de la conférence