Prediction of thermal conductivity of liquid and vapor refrigerants for pure and their binary, ternary mixtures using artificial neural network - Université de Rennes Accéder directement au contenu
Article Dans Une Revue Thermophysics and Aeromechanics Année : 2019

Prediction of thermal conductivity of liquid and vapor refrigerants for pure and their binary, ternary mixtures using artificial neural network

Résumé

The determination of thermophysical properties of hydrofluorocarbons (HFCS) is very important, especially the thermal conductivity. The present work investigated the potential of an artificial neural network (ANN) model to correlate the thermal conductivity of (HFCS) at (169.87-533.02) K, (0.047-68.201) MPa, and (0.0089-0.1984) W/(m·K) temperature, pressure, and thermal conductivity ranges, respectively, of 11 systems from 3 different categories including five pure systems (R32, R125, R134a, R152a, R143a), four binary mixtures systems (R32 + R125, R32 + R134a, R125 + R134a, R125 + R143a), and two ternary mixtures systems (R32 + R125 + R134a, R125 + R134a + R143a). Each one received 1817, 794 and 616 data points, respectively. The application of this model for these 3227 data points of liquid and vapor at several temperatures and pressures allowed to train, validate and test the model. This study showed that ANN models represent a good alternative to estimate the thermal conductivity of different refrigerant systems with a good accuracy. The squared correlation coefficients of thermal conductivity predicted by ANN were R2 = 0.998 with an acceptable level of accuracy of RMSE = 0.0035 and AAD = 0.002 %. The results of applying the trained neural network model to the test data indicate that the method has a highly significant prediction capability.

Domaines

Chimie
Fichier principal
Vignette du fichier
Ghalem et al-2019-Prediction of thermal conductivity of liquid and vapor refrigerants.pdf (1.93 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02515141 , version 1 (23-03-2020)

Identifiants

Citer

N. Ghalem, S. Hanini, Mx Naceur, M Laidi, Abdeltif Amrane. Prediction of thermal conductivity of liquid and vapor refrigerants for pure and their binary, ternary mixtures using artificial neural network. Thermophysics and Aeromechanics, 2019, 26 (4), pp.561-579. ⟨10.1134/S0869864319040085⟩. ⟨hal-02515141⟩
22 Consultations
309 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More