Development of a Data-Driven Approach for Human-Based Environmental Control

Abstract

The conventional HVAC industry control method has relied on the Predicted Mean Vote (PMV) model [1]. However, the PMV model has frequently met a technical drawback in satisfying a building occupant’s thermal comfort requirements since there are several limitations and assumptions inherent in the PMV model. That model cannot be considered because of differences in individual occupant’s non-linear features, such as age, gender, and race, as well as the Body Mass Index (BMI), that adversely impact the thermal comfort in PMV model-based control. Artificial intelligence algorithm was applied to the thermal comfort control, which captured each individual occupant’s thermal preference. Human thermal comfort experiments were conducted to collect thermal comfort preference data on each occupant, and validation experiments tested the performance of the artificial intelligence algorithm. The study results indicated that the artificial intelligent algorithm based model had up to 45% more energy savings and 44.3% better thermal comfort performance than the PMV based model.