PREVIZIONAREA PREȚULUI ELECTRICITĂȚII ÎN BAZA DATELOR DE ÎNCĂRCARE

Abstract

Accurate forecasting of electricity prices and tariffs is essential to maintain a stable interaction between supply and demand in the dynamic electricity market. The paper describes a convolutional neural network-based model for the day-ahead electricity price forecasting from historical prices / loads and predicted values of the load. The model was tested on the data for New York and New South Wales and achieved high prediction accuracy on both data sets.