DERİN ÖĞRENME TEKNİKLERİYLE TÜRKİYE GÜN ÖNCESİ PİYASASINDA ELEKTRİK FİYAT TAHMİNİ

Day-Ahead Market offers electricity market participants the opportunity to trade one day ahead of real-time. For each hour, a separate Clearing Price is created in Market. This study aims predict hourly using deep learning techniques. In this context, 24-hour Prices were forecasted with MLP, CNN, LSTM, and GRU. LSTM had best average forecasting performance an 8.15 MAPE value, according results obtained. MLP followed 8.44 MAPE, GRU 8.72 CNN 9.27 MAPE. study, provinces where power plants producing renewable resources are dense selected for meteorological variables. It expected that trend towards generation will increase future. it thought important consider factors may affect production these price forecasting.