Forecasting Accuracy Improvement of Solar Power Plant's Generation

Forecasting RES generation is an important task both in finding the optimal location for newly commissioned RES generation facilities, and in solving problems of existing electric power systems with a high share of RES. Therefore, obtaining high accuracy of forecasting results is one of the main conditions for the successful development of renewable energy sources. This paper studies ways to improve the accuracy of forecasting the solar power plant generation using machine learning methods. The paper describes the application of preprocessing and preparation of initial data on the forecasting accuracy. The paper also analyzes the application of three different regression models to solving the forecasting problem and gives an accuracy evaluation of each of the models.