LikeSNU-서울대 중앙도서관 빅데이터 지식 플랫폼

비전형 잔여 기상 정보를 활용한 하이브리드형 전력 수요 예측 모델 설계

Author

박강구, 송준호, 황의석

Journal Title

정보과학회 컴퓨팅의 실제 논문지

Publication Year

2019

Topics

#Hybrid
#Demand forecasting

Summary

This study designs a hybrid power demand forecasting model combining linear and nonlinear prediction approaches for complex electricity usage data influenced by meteorological information and human factors, and proposes a method to identify key input variables of the nonlinear model based on random forest. The prediction model considering atypical residue of meteorological information demonstrates improved prediction reliability.

Go Source Add to Selection