영상이미지 기반 착의량 산출 모델 적용에 따른 실내 PMV 제어 및 열쾌적 평가

Abstract

Purpose: This study aims at evaluating the thermal comfort of occupants using a PMV (predicted mean vote)-based HVAC (Heating, ventilation, and air conditioning) controls with dynamic clothing insulation estimated from occupant image. The contribution of this study is to highlight 1) the estimation method for the insulation of clothing ensembles that can appear in various ways inside a building, and 2) the effect of PMV-based control reflecting dynamic clothing insulation on the actual thermal comfort. Method: For the comparison, three control methods (Mode 1-3) including the typical control methods were applied and analyzed in terms of the room temperature, PMV, and subjects’ TSV (Thermal sensation vote). For the experiment, a Test-bed equipped with environmental sensors and camera sensors was built. A total of 6 subjects were experimented for 40 minutes per one clothing ensemble. Result: Although estimating error of the dynamic clothing insulation occurred in some clothes of the female subjects, it was confirmed that, nevertheless, the PMV-based control reflecting the dynamic clothing insulation (Mode 3) was able to create a more comfortable environment than other control methods. Therefore, this study emphasize the potential of promoting customized comfort environment to the occupant through vision-based dynamic clothing insulation.