The Directive 2009/28/EC of the European Parliament and of the Council sets a national overall target of 13 percent for the share of energy from renewable sources in gross final consumption of energy to be reached by 2020. To meet this obligation, the currently installed wind capacity of 330 MW is expected to increase. However, integration of large amount of wind capacity into the grid introduces several challenges.
Because of the stochastic nature of wind, the operation of wind power plants and conventional generation units differ remarkably from each other. Wind power output is highly variable at several time scales and although state-of-the-art forecasting methods are used, the accuracy of wind power prediction falls behind when compared to load forecast.
The aim of this thesis is to review the issues related to ramp events caused by sudden changes in wind speed, and the benefits of geographical dispersion. As the variations of wind turbine output spread out over a large area containing several wind farms, the aggregated power output is smoothed.
For the statistical analysis of ramp events, measured data from individual wind power plants situated in Hungary is used. The nameplate capacity of the investigated plants ranges from 0,6 MW to 44 MW and the distance between them varies between 0 and 325 km. Power ramps are calculated for 3 different time horizons and are integrated over the area in order to compare the measures of power production variability of a single plant to the variation of the aggregated output. For further comparison, correlation of power ramps in relation to nominal capacity and distance is also investigated.