TY - GEN
T1 - A Short-Term Electric Load Forecasting Scheme Using 2-Stage Predictive Analytics
AU - Moon, Jihoon
AU - Kim, Kyu Hyung
AU - Kim, Yongsung
AU - Hwang, Eenjun
N1 - Funding Information:
ACKNOWLEDGMENT This work was supported by the Korea Institute of Energy Technology Evaluation and Planning (KETEP) and the Ministry of Trade, Industry & Energy (MOTIE) of the Republic of Korea (No. 20152010103060) and Institute for Information & communications Technology Promotion (IITP) grant funded by the Korea government (MSIT) (No. R0190-16-2012, High Performance Big Data Analytics Platform Performance Acceleration Technologies Development).
Publisher Copyright:
© 2018 IEEE.
PY - 2018/5/25
Y1 - 2018/5/25
N2 - One key issue for stable power supply is to forecast electric load accurately. Since buildings of the same type show similar power consumption patterns, it should be considered for accurate electric load forecast. In particular, university buildings show various electric loads depending on time and other external factors. In this paper, we propose a short-term load forecast model for educational buildings using 2-stage predictive analytics for the effective operation of their power system. To do that, we collect the electric load data of five years from a university campus. Next, we consider the electric load pattern by using the moving average method according to the day of the week. Next, we predict the daily electric load using the random forest method and finally evaluate its performance using the time series cross-validation. The experimental results show that our forecasting model outperforms other competing methods in terms of prediction accuracy.
AB - One key issue for stable power supply is to forecast electric load accurately. Since buildings of the same type show similar power consumption patterns, it should be considered for accurate electric load forecast. In particular, university buildings show various electric loads depending on time and other external factors. In this paper, we propose a short-term load forecast model for educational buildings using 2-stage predictive analytics for the effective operation of their power system. To do that, we collect the electric load data of five years from a university campus. Next, we consider the electric load pattern by using the moving average method according to the day of the week. Next, we predict the daily electric load using the random forest method and finally evaluate its performance using the time series cross-validation. The experimental results show that our forecasting model outperforms other competing methods in terms of prediction accuracy.
KW - Electric Load Forecasting
KW - Forecasting Model
KW - Machine Learning
KW - Moving Average
KW - Random Forest
UR - http://www.scopus.com/inward/record.url?scp=85048461574&partnerID=8YFLogxK
U2 - 10.1109/BigComp.2018.00040
DO - 10.1109/BigComp.2018.00040
M3 - Conference contribution
AN - SCOPUS:85048461574
T3 - Proceedings - 2018 IEEE International Conference on Big Data and Smart Computing, BigComp 2018
SP - 219
EP - 226
BT - Proceedings - 2018 IEEE International Conference on Big Data and Smart Computing, BigComp 2018
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2018 IEEE International Conference on Big Data and Smart Computing, BigComp 2018
Y2 - 15 January 2018 through 18 January 2018
ER -