Key points are not available for this paper at this time.
This research aimed to develop methods for using Big Data technologies to forecast electricity generation from solar photovoltaic power plants, which is crucial for optimising energy production and increasing the efficiency of solar resource utilisation. The study employed a method of analysing the economic feasibility of using energy storage systems and a comparative analysis of electricity buying and selling prices on the market. An experiment involving software tools and algorithms for processing, analysing, and modelling large volumes of data was also conducted. As a result of the research, methodologies were developed that encompass data collection and analysis, information visualisation, selection and training of forecasting models based on available data, as well as monitoring and testing their effectiveness. Graphical diagrams were constructed to illustrate the stages of data processing and analysis, the process of forecasting electricity generation for different time periods, and the process of training a model based on data, monitoring, and testing the model. Additionally, a graph was created to show the typicality and range of values, and a graph to display the change in electricity prices throughout the day. Furthermore, technological tools for using Big Data were described, the cost of electricity was calculated, and the economic attractiveness of using energy storage systems was assessed. As a result of the research, a potential profit indicator from price arbitrage was established, as well as economic parameters for the feasibility of using energy storage management based on an analysis of differences in electricity purchase and sale prices. The results obtained can be useful for energy companies and organisations involved in the production of electricity from solar photovoltaic power plants, allowing them to optimise energy production and increase the efficiency of solar resource utilisation
Oleksandr Stoliarov (Mon,) studied this question.