Model for Forecasting and Planning Solar Energy Production Using an Artificial Neural Network

The aim of this paper is to investigate and improve existing approaches for efficient positioning of solar power generation facilities and a model for short-term forecasting of the generated energy of photovoltaic systems, using a statistical method and neural network training, to effectively locate the generation facilities. The proposed approach allows to obtain an accurate and justified model for deploying photovoltaic systems to solve the problem of active power reserve. The algorithm for short-term forecasting is built on the basis of the multiple regression method, which allows to forecast for a generation with an error of no more than 20%.