| Title |
A Deep Learning Model and Training Technique for 240 Hours Load Forecasting in Korea Power System |
| Authors |
문찬호(Chan-Ho Moon) ; 권보성(Bo-Sung Kwon) ; 송경빈(Kyung-Bin Song) |
| DOI |
https://doi.org/10.5370/KIEE.2022.71.4.585 |
| Keywords |
240 Hours Load Forecasting; Deep Learning; Training Technique; Weather Prediction; Time Intervals |
| Abstract |
It is essential to forecast 240 hours load accurately for stable power system operation in South Korea. Training technique for 240 hours load forecasting using deep learning is proposed. Suitable training technique for 240 hours load forecasting is developed using deep learning model. The 240 hours load forecasting method is proposed applying training technique which consists of two different types for weather prediction. According to usable weather factors depending on forecast ranges, training technique is designed by reflecting time intervals between forecasting time and forecasting target. The proposed method has improved prediction accuracy compared to the existing exponential weighted moving average method |