Selected aspects of generating short-term electricity demand forecasts, taking into account renewable energy sources

Abstract

The risk of a decline in the quality of electricity demand forecasts in the short term increases due to the increase in the installed capacity of renewable energy sources (RES). This is mainly due to the high daily variability of electricity production from renewable sources, which is strongly dependent on local weather conditions. Production from renewable energy sources is a very complex time series, additionally reinforced by a significant increase in its share in total production. This applies in particular to photovoltaic sources in low-voltage networks. There is therefore an urgent need to improve the quality of forecasts in this area. The main goal of the research was to verify statistical models that often achieve good results in the complex problem of forecasting electricity demand. The main objective, regarding daily forecasts of consumers' demand for electricity, was achieved through the implementation of intermediate objectives, including the development of a methodology for estimating electricity generated by photovoltaic installations.