Partager

Recommandé par l'IIF

Recherche d’un modèle de pré-refroidissement économe en énergie pour un climatiseur à variation de fréquence grâce à un algorithme d’apprentissage par renforcement approfondi.

Energy saving pre-cooling pattern search of an inverter air conditioner using a deep reinforcement learning algorithm.

Numéro : No 339

Auteurs : YOON M. S., YOON W. S.

Résumé

In this study, we experiment with the energy saving pre-cooling operation of an inverter air conditioner using a deep reinforcement learning algorithm under a fixed going out time duration. An air conditioner with a rated cooling capacity of 6500 W is used. An indoor building load of 4500 W sensible heat and an outdoor temperature of 35°C are continuously applied throughout the study. During the going out time, three remote control temperature options (21, 26, 31°C) are given for simplicity, and a remote controlling pattern (set temperature sequences) that minimizes power consumption upon arrival home while satisfying the arrival target temperature condition is searched using deep reinforcement learning. For a given going out time duration, the power consumptions of the temperature operation pattern found by the computer deep reinforcement learning algorithm and a human are compared. Although the artificial intelligence searches for the power consumption operation pattern over a long period (7 ~ 9 days), depending on the learning environment, it identifies a more accurate and more energy saving and an unexpected operation pattern than by human analysis. 

Documents disponibles

Format PDF

Pages : 12 p.

Disponible

Gratuit

Télécharger

Détails

  • Titre original : Energy saving pre-cooling pattern search of an inverter air conditioner using a deep reinforcement learning algorithm.
  • Identifiant de la fiche : 30030108
  • Langues : Anglais
  • Sujet : Technologie
  • Source : 13th IEA Heat Pump Conference 2021: Heat Pumps – Mission for the Green World. Conference proceedings [full papers]
  • Date d'édition : 31/08/2021

Liens


Voir d'autres communications du même compte rendu (198)
Voir le compte rendu de la conférence

Indexation