[3Xin2] Poster session 1

[3Xin2-68] A basic study of electricity demand forecasting using cloud cover data of satellite imagery

〇Riku Niizawa¹, Shoichi Urano¹ (1.Meiji University)

Keywords:Machine Lerning, Electricity demand forecasting, Satellite imagery, Cloud cover

The demand for electricity is closely related to people's behavior, and therefore, weather information is often utilized in electricity demand forecasting. The primary weather information commonly utilized is temperature and humidity, and the effectiveness of these variables have been clearly demonstrated. Therefore, in this paper, satellite image data is employed to analyze the sky's conditions, which significantly influence people's behavior. In addition to temperature and humidity, this paper aims to obtain cloud cover from satellite imagery as a basic study, and to use it as a characteristic quantity for electricity demand forecasting to improve the accuracy of electricity demand forecasting one hour ahead and the next day.