PREDVIĐANJE CENA ELEKTRIČNE ENERGIJE NA BERZI PRIMENOM MAŠINSKOG UČENJA

In this paper, we present machine learning models that were developed in order to predict the price on the electricity market in Serbia. Various algorithms will be presented with an analysis of their results. The models presented are based on the principles of models used to predict time series and a complex machine learning model based on relevant data that affects the price of electricity in Serbia. We will examine to what extent such modern algorithms can be applied to the problem of price forecasting in the electricity market. Historical data taken from the transparency platform of the Association of European Transmission System Operators (https://transparency.entsoe.eu/) will be used to create the model and analysis. The algorithm and analysis were made in the programming language "Python" using appropriate libraries. The algorithm with the best results will be presented and publicly available as an Internet service that displays the model's day-ahead predictions, as well as historical model data to check the quality of the predictions.