| Keywords |
Feature importance; Correlation analysis; Weather and time data; Cooling electricity consumption; Commercial buildings |
| Abstract |
Building sector takes responsible for approximately 30% of the national energy consumption in Korea. For energy monitoring and management, highly energy usage buildings which have the large service area and the high energy-use-intensity have been regulated to report the energy consumption data to the public since 2008. Although the energy data has been applicable to provide various information on the energy performance of the building, it has necessarily to characterizes the correlation between the building energy consumption and parameters which affects on the energy usage. This research collected hourly electricity consumption data of ten large commercial buildings among the national highly energy usage building list and investigated the impact of each parameters on the electricity consumption under different time resolutions, hourly and daily in summer. The results shows that daily mean outdoor air temperature is the most correlated variable and feature compared with other variables to illustrate the daily energy consumption but the larger building area is, the less R2 values between the temperature and electricity consumption.
In addition, the coefficient of the time flag is comparable to hourly outdoor air temperature under the correlation analysis but the parameter is the most importance feature for hourly electricity energy consumption. |