Short-Term Forecasting of Photovoltaic Solar Power Generation Based on Time Series: Application for Ensure the Efficient Operation of the Integrated Energy System of Ukraine

Abstract

Over the last decade, there has been a growing in the dependence of electricity production by solar power plants (SPPs) in Ukraine. Therefore, there is a need to optimize the structure of the energy balance of the state, based on the requirements of energy security and ensure the share of renewable energy at 25%. However, with the development of renewable energy sources (RESs) there is a problem of ensuring the appropriate maneuverability of the power system. This is due to the fact that the structure of generating capacity of the United Power System of Ukraine in terms of effective regulation of frequency and power in the power system is suboptimal. Among the reasons for this, the main ones are unregulated and variable operation of a SPP, which is exacerbated by the lack of tools and approaches for the power generation modes forecasting. That is why the issue of accurate forecasting of the possible electricity generation volume has become acute. However, solar energy forecasting is a rather difficult task, as it largely depends on climatic conditions that change over time. This study presents an analysis and application of the seasonal autoregressive integrated moving average (SARIMA) method to develop a model that can support and provide forecasting the amount of power produced by SPP. Data for the development of the model were obtained from the time series of electricity generation on the example of the SPP in the village of Velyka Dymerka, Kyiv re-gion. The data consisted of more than 26 thousand samples collected from July 1, 2020, to December 31, 2020, which characterize the operating conditions of solar panels with a capacity of 9 MW. This led to the choice of the SARIMA model. The coefficient of determination (R2) for the obtained model was 92%. This indicates the ability of the final model to accurately represent and give forecast based on data set of the SPP power generation.