Stochastic Simulation of Wind Power Profiles from Time Series Analysis Considering Dependencies on Meteorological Variables

Due to the higher and higher shares of generation from variable renewable energy sources, electric power systems are characterized by increasing variability and uncertainty that call for the application of probabilistic methods at different stages of system management. For the evaluation of adequacy and techno-economic indices within the scope of grid planning applications, Monte Carlo simulation is a typical approach, whose iterations are fed by instances of time series of stochastic quantities such as load demand and renewable production. Here the method underlying “SPOPSI_wind” (Stochastic wind POwer Profile SImulator) is described and validated, that starting from the analysis of historical series of wind power in Italian regions and considering the dependence of historical wind power on temperature and wind speed, synthesizes a model that is used to generate new plausible stochastic series in the future. The new series maintain the statistical properties of the past but exhibit a significant variability, thus being suitable for simulating a wide range of plausible power system operating conditions. Moreover, the formulation has the potential to deal with the impact of climate changes as well.