Partager

Document IIF

Commande prédictive des colonnes à pulvérisation dans les mines souterraines : approche intégrant l’apprentissage automatique et l’Internet des objets.

Predictive control of underground mine spray chambers: An integrated machine learning and IoT approach.

Auteurs : PANDEY A., SINGH S. K., SRIDHARAN S. J., MISHRA S.

Type d'article : Article de la RIF

Résumé

This study presents an innovative Internet of Things (IoT) solution for maintaining constant air temperature at the outlet of spray chambers in underground coal mines. The system addresses the challenge of varying inlet air conditions caused by diurnal and seasonal fluctuations, which significantly impact mine ventilation efficiency and worker comfort. The proposed IoT architecture integrates multiple sensors to monitor key parameters including inlet air temperature and pressure, chilled water temperature, and air mass flow rate. A central control unit processes this real-time data to dynamically adjust the chilled water flow rate, ensuring consistent air temperature regardless of inlet condition variations. The system’s design is based on extensive psychrometric data collected from an underground coal mine, with inlet conditions extrapolated for various depths using specialized software. Nine machine learning algorithms were evaluated to predict optimal chilled water flow rates, with XGBoost demonstrating superior performance (R² = 0.994, RMSE = 0.566). Feature importance analysis revealed that inlet air sigma heat, desired outlet air temperature, and spray chamber factor of merit were the most influential parameters affecting water flow requirements. Through continuous optimization of chilled water usage, this intelligent system maintains desired air conditions while promising substantial energy savings compared to traditional fixed-setting spray chambers. The proposed three-layer IoT architecture, comprising device, edge computing, and network layers, enables real-time monitoring and adaptive control. This approach to mine ventilation represents a significant advancement in mining technology, potentially improving both operational efficiency and miners’ working conditions.

Documents disponibles

Format PDF

Pages : 14 p.

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 : Predictive control of underground mine spray chambers: An integrated machine learning and IoT approach.
  • Identifiant de la fiche : 30034372
  • Langues : Anglais
  • Sujet : Technologie
  • Source : International Journal of Refrigeration - Revue Internationale du Froid - vol. 179
  • Date d'édition : 11/2025
  • DOI : http://dx.doi.org/https://doi.org/10.1016/j.ijrefrig.2025.08.009

Liens


Voir d'autres articles du même numéro (29)
Voir la source

Indexation