Partager

Document IIF

Prédictions de la rétention d’huile dans les conduites horizontales et verticales de vapeur de frigorigène des systèmes split unitaires à l’aide d’un modèle basé sur l’apprentissage machine fondé sur la physique.

Predictions of oil retention in horizontal and vertical refrigerant vapor lines of unitary split systems using a physics-based machine learning-aided model.

Numéro : 0405

Auteurs : BAHMAN A. M., ERFANIMATIN M., NOURANI P., LIU H., SHAH V. M., BRAUN J. E., GROLL E. A.

Résumé

Due to deployment of variable-speed unitary air conditioning (AC) systems as well as the future implementation of newer HFO refrigerants, there is a need to upgrade line sizing guidelines to properly account for the effects of oil retention (OR). These new guidelines can be facilitated through the development and application of a model that can predict the OR in gas lines for commonly used refrigerant-lubricant combinations in the HVAC&R industry. This work aims to evaluate the prediction performance of machine learning (ML) models that are trained using OR data obtained for horizontal and vertical lines of different diameters (11, 17, 20 mm), for different refrigerants (R-134a, R-410A, R-32, and R-1234ze(E)) mixed with POE-32 lubricant at different flow conditions. The results of the predictions from the ML models are compared to an analytical model. The ML models were trained using experimentally collected data based on more than 240 tests. The inputs to each model are refrigerant conditions (type, temperature, pressure, and mass flow rate), pipeline dimension and orientation, injected oil mass flow rate, and viscosity, whereas the model output is OR. Several model types were investigated to predict the OR that included a physics-based model, two standalone ML models, and two physics-based machine learning-aided (PBMLA) algorithms. The results showed that the standalone ML algorithms performed poorly for OR prediction compared to the analytical model. The results also showed that the PBMLA models could predict OR slightly better than the analytical model.

Documents disponibles

Format PDF

Pages : 10

Disponible

  • Prix public

    20 €

  • Prix membre*

    Gratuit

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

Détails

  • Titre original : Predictions of oil retention in horizontal and vertical refrigerant vapor lines of unitary split systems using a physics-based machine learning-aided model.
  • Identifiant de la fiche : 30031513
  • Langues : Anglais
  • Sujet : Technologie
  • Source : Proceedings of the 26th IIR International Congress of Refrigeration: Paris , France, August 21-25, 2023.
  • Date d'édition : 21/08/2023
  • DOI : http://dx.doi.org/10.18462/iir.icr.2023.0405

Liens


Voir d'autres communications du même compte rendu (492)
Voir le compte rendu de la conférence

Indexation