Partager

Prévision de la performance d'une pompe à chaleur couplée sol-eau utilisant des réseaux neuronaux avec des données statistiques pondérant le prétraitement.

Forecasting of a ground-coupled heat pump performance using neural networks with statistical data weighting pre-processing.

Auteurs : ESEN H., INALLI M., SENGUR A., et al.

Type d'article : Article

Résumé

The objective of this work is to improve the performance of an artificial neural network (ANN) with a statistical weighted pre-processing (SWP) method to learn to predict ground source heat pump systems with the minimum data set. Experimental studies were completed to obtain training and test data. Air temperatures entering/leaving condenser unit, water-antifreeze solution entering/leaving the horizontal ground heat exchangers and ground temperatures (1 and 2 m) were used as input layer, while the output is COP of system. Some statistical methods, such as the root-mean squared (RMS), the coefficient of multiple determinations (R2) and the coefficient of variation is used to compare predicted and actual values for model validation. It is found that RMS value is 0.074, R2 value is 0.9999 and cov value is 2.22 for SCG6 algorithm of only ANN structure. It is also found that RMS value is 0.002, R2 value is 0.9999 and cov value is 0.076 for SCG6 algorithm of SWP-ANN structure. The simulation results show that the SWP based networks can be used an alternative way in these systems. Therefore, instead of limited experimental data found in literature, faster and simpler solutions are obtained using hybridized structures such as SWP-ANN. [Reprinted with permission from Elsevier. Copyright, 2007].

Détails

  • Titre original : Forecasting of a ground-coupled heat pump performance using neural networks with statistical data weighting pre-processing.
  • Identifiant de la fiche : 2008-1790
  • Langues : Anglais
  • Source : International Journal of thermal Sciences - vol. 47 - n. 4
  • Date d'édition : 04/2008

Liens


Voir d'autres articles du même numéro (1)
Voir la source

Indexation