Publications

Incorporating Phase Change Materials (PCMs) into indoor environments can enhance users' thermal comfort and improve the building's  energy  efficiency. However, the number of PCM studies in hot and humid climates, such as most Brazilian conditions, is comparatively less developed than research for temperate climates. This article investigates the influence of climatic  parameters  on  the  PCMs'  function  to  enhance  user  thermal  comfort  in  an  office  model  in  Brazilian  climates employing  thermo-energy  simulation  using  EnergyPlus  for  95  cities  with  adaptive  thermal  comfort  as  a  performance indicator. We employed machine learning techniques, including Gradient Boosting and Feature Importance, to quantify the influence  of  climate  parameters,  followed  by  Principal  Component  Analysis  clustering.  Results  show  that  incorporating PCMs can potentially increase user thermal comfort in part of Brazil, mainly in cities with average temperatures below 21°C. The comfort increase is not homogeneous, being more relevant in regions  with colder temperatures and higher altitudes. When  considering  better-performing  cities,  the  most  influential  climatic  parameters  were  dry  bulb  temperature,  relative humidity, and wind direction.

Keywords: phase change material, adaptive thermal comfort, EnergyPlus, climate parameters, machine learning.