Statistical processing of wind and solar PV generation variability for assessment of additional power system flexibility

Abstract

To solve the actual task of assessing additional flexible generation needed to smooth variable renewable energies (VRE) disturbances in the Power Systems worldwide the original method for statistical processing of wind and solar PV generation variability data was proposed in the article. The proposed method allows processing the historical data of VRE day-by-day and hour-to-hour generation and assessing the levels of potential disturbances this type of generation could cause. These disturbances should be smoothened in the Power System by others, usually conventional generators at least to ensure grid stability and electricity quality. Currently, at the stage of the power systems starts transformation, the sources of power system flexibility basically are the power plants whose individual flexibility is characterized by three dimensions: adjustability, ramping and lead time. Lack of flexible generation causes VRE curtailments that negatively affect economic efficiency of such clean generation. From the other hand, the improvement of existing power plants flexibility requires an essential investments and time to retrofit them, and because of that the important question - what is the required level of power system flexibility is needed to smooth current and future VRE generation. The results of statistical data processing for wind and solar PV hour-by-hour power output changes allow accepting the hypothesis about the normal distribution of these data sets. The article presents the key parameters of theoretical normal distributions are calculated for wind and solar PV data sets for 2019 and 2020 as a whole, and also for winter and summer seasons, which allow obtaining a more precise assessment of variability levels for renewables, especially for solar PV hour-by-hour generation changes. The assessments obtained in the article also could be used as the first step for more efficient management of the electricity market, for instance, forecasting the required level of flexible generation or ancillary services market for intraday power system balancing. Keywords: variable renewable energy, power system, flexibility, generation profile, statistical data processing