ZEMCH 2019 International Conference Proceedings April.2020 | Page 166
1. Introduction
There has been an increasing demand for indoor comfort control to raise occupant satisfaction and
work efficiency by ensuring comfort levels [1,2]. The most widely used method utilizes the thermal
environment indicator Fanger PMV (Predicted Mean Vote) [3]. However, there is a limit to measuring
the accurate TSV (Thermal Sensation Vote) through the Fanger PMV derived through chamber
experiments [4]. Various studies carried out to improve the prediction capacity of Fanger PMV [5‐7].
Besides the thermal environment, indoor air quality is also an important factor which affects the
comfort levels in a room. If the indoor air quality is not secured, the occupant may feel that it is difficult
to breathe or suffer from harmful health effects [8,9]. The PMV comfort control method does not take
into account such air quality aspects and considers only the thermal environment, limiting its capacity
to satisfy occupant comfort. Previous studies have looked at the correlation between PMV and carbon
dioxide levels—a representative indicator of air quality. Lee et al. have shown that PMV value and
carbon dioxide concentration have a high correlation in an office setting with mechanical ventilation
[10]. Kim et al. showed that when developing a PMV prediction model using ANN, adding carbon
dioxide as a variable reduced the errors although the difference was slight [11]. As such, previous
studies have looked at the relations between indoor air quality and the thermal environment, but an
index that considers both the conditions is yet to be developed. Hence, as an initial step to develop a
comfort indicator that integrates indoor air quality and the thermal environment, this study analyzed
the impact of carbon dioxide concentration on TSV by taking measurements of the indoor environment
and collecting TSV data.
2. Methods
2.1. Measurement
The target space in this study was reading room #4 at S university located in Seoul, South Korea.
where an individual air conditioning system is used without separate ventilation facilities. The room is
open from 6 to 12 am, as well as being open for 24 h every other day. The reading room uses an
individual HVAC system, which operates continuously, with no dedicated mechanical ventilation
system. As the set‐point temperature is set at the discretion of the room manager, we were unable to
collect data regarding the set‐point temperature of the air conditioner. A reading room of a university
library has diverse occupant density and carbon dioxide levels according to time, making it an adequate
location to carry out the study.
Four physical indicators were collected as seen in Table 1 to calculate the PMV. A thermal
environment measurement devices were installed in the center of the room to measure the mean radiant
temperature, indoor air temperature, and indoor relative humidity, and two additional measurement
devices were installed to measure the temperature and humidity near the windows. The metabolic rate
and clo levels were set at 1.0 met and 0.75 clo, and the air velocity was assumed to be 0.1 m/s. To
measure from the actual location where the occupants breathe sitting in their chairs, the carbon dioxide
concentration measurement devices were installed at two locations at the center of the room at the
height of 1 m. The number of occupants was collected hourly using the library app, and the collected
data was used to analyze its correlation with carbon dioxide levels.
Table 1. Measurement Overview
Location S university library reading room
(Capacity: 264)
Term 2019/03/22~2019/04/30
(3/22~4/7: Non Examination, 4/8~4/30: Examination)
Elements & Instruments
155
Wet bulb globe Temp.
FPA805GTS
ZEMCH 2019 International Conference l Seoul, Korea