Document IIF

Modélisation d'un système à pompe à chaleur sol-eau à l'aide de systèmes neuronaux et logique floue par inférence.

Modelling a ground-coupled heat pump system using adaptive neuro-fuzzy inference systems.

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

Type d'article : Article, Article de la RIF


The aim of this study is to demonstrate the usefulness of an adaptive neuro-fuzzy inference system (ANFIS) for the modelling of ground-coupled heat pump (GCHP) system. The GCHP system connected to a test room with 16.24 m2 floor area in Firat University, Elazig. (38.41°N, 39.14°E), Turkey, was designed and constructed. The heating and cooling loads of the test room were 2.5 and 3.1 kW at design conditions, respectively. The system was commissioned in November 2002 and the performance tests have been carried out since then. The average performance coefficients of the system for horizontal ground heat exchanger in the different trenches, at 1 and 2 m depths, were obtained to be 2.92 and 3.2, respectively. Experimental performances were performed to verify the results from the ANFIS approach. In order to achieve the optimal result, several computer simulations have been carried out with different membership functions and various number of membership functions. The most suitable membership function and number of membership functions are found as Gauss and 2, respectively. For this number level, after the training, it is found that root-mean squared value is 0.0047, and absolute fraction of variance value is 0.9999 and coefficient of variation in percent value is 0.1363. This paper shows that the values predicted with the ANFIS, especially with the hybrid learning algorithm, can be used to predict the performance of the GCHP system quite accurately.

Documents disponibles

Format PDF

Pages : 65-74


  • Prix public

    20 €

  • Prix membre*


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


  • Titre original : Modelling a ground-coupled heat pump system using adaptive neuro-fuzzy inference systems.
  • Identifiant de la fiche : 2008-0878
  • Langues : Anglais
  • Source : International Journal of Refrigeration - Revue Internationale du Froid - vol. 31 - n. 1
  • Date d'édition : 01/2008


Voir d'autres articles du même numéro (16)
Voir la source