| Abstract | As part of efforts to achieve carbon neutrality, improving building energy performance has become a key objective, with a focus on green remodeling and enhancing energy efficiency. However, existing approaches face limitations due to restricted data collection and discrepancies between standard thermal indices like PMV and actual occupant thermal preferences. To address these issues, this study develops an algorithm to derive occupant comfort ranges based on indoor/outdoor environmental data and cooling system control history. Using elimination, Artificial Neural Networks (ANN), and K-Nearest Neighbor (KNN) algorithms, the study defines comfort and discomfort points and derives the optimal comfort range. Energy consumption simulations were conducted using specific points within this range. The proposed methodology improves as control data accumulates, offering a strategy to ensure comfort while reducing energy consumption. |
|---|
