A long-term perspective of wind power output variability

Abstract

The study of the wind power output variability comprises different time scales. Among these, low-frequency variations can substantially modify the performance of a wind power plant during its lifetime. In recent years, other temporal scales such as the short-term variability or the climatological conditions of wind and the corresponding generated power have been investigated in depth. However, the study of longer decadal and multidecadal variations is still in its early stages. In this work, the wind power output long-term variability is analysed for two locations in Spain, during the period 1871-2009. This is attained by computing the annual wind speed probability density functions derived from an ensemble of atmospheric sea level pressure data set through a statistical downscaling based in evolutionary algorithms. Results reveal significant trends and periodicities in multidecadal bands including 13, 25 and 46years, as well as significant differences among both sites. The impact of the leading large-scale circulation patterns (NAO, EA, SCAND and AMO) on wind power output and its stationarity is analysed. Results on both locations show non-stationary significant and opposite seasonal couplings with these forcings. Finally, the long-term variability of the reconstructed Weibull parameters of the annual wind speed distributions is used to derive a linear model to estimate the annual wind power.