Analysis of Thermal Distribution Patterns in Multi-Occupancy Environments


Topak F., Pavlak G. S., Pekeriçli M. K., Wang J.

ASHRAE 2023 Annual Conference, Florida, Amerika Birleşik Devletleri, 24 - 28 Haziran 2023, cilt.129, ss.445-452

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Cilt numarası: 129
  • Doi Numarası: 10.13140/rg.2.2.33941.70883
  • Basıldığı Şehir: Florida
  • Basıldığı Ülke: Amerika Birleşik Devletleri
  • Sayfa Sayıları: ss.445-452
  • Orta Doğu Teknik Üniversitesi Adresli: Evet

Özet

Despite the large share of heating, ventilation, and air conditioning (HVAC) systems in building energy use, the lack of thermal comfort is still considered as one of the unresolved issues for building occupants. While recent research efforts on individualized comfort assessment and personalized control systems have shown promising insights for maintaining occupant comfort in single occupancy spaces, generating collectively acceptable conditions in multi-occupancy indoor environments where occupants with varying comfort profiles share the same thermal zones is still an outstanding question. Current studies in this field are generally built upon the assumption that the zone is well-mixed and temperature is uniform throughout, which ignores the dynamic conditions of particular locations. Very limited research exists questioning the potential of incorporating micro-thermal condition data in control loops for system efficiency in buildings. However, accounting for the heterogeneity of indoor climate conditions holds great potential for multi-occupancy indoor spaces, given the differences in the personal comfort preferences of occupants. To this end, this research aims to analyze thermal distribution patterns in multi-occupancy environments to support occupant-centric control strategies. Computational fluid dynamics (CFD) was used to model and simulate thermal conditions variations in an office space shared by six occupants under different scenarios with varying airflow rates, airflow directions, and occupancy conditions. The thermal distribution dataset compiled with CFD simulations was analyzed using the Random Forest algorithm to evaluate the importance of simulation parameters. Our results showed that the thermal conditions to which occupants are subjected vary considerably based on their particular locations, and temperature distribution patterns could be influenced by controlling the direction and flow rate of the supplied air. The temperature values at occupant locations differ up to two and a half degrees Celsius (~five degrees Fahrenheit) in different cases with constant supply airflow settings, demonstrating the importance of micro-climate consideration in thermal comfort assessments.