A comparative metric to quantify the variability of wind power

The stochastic nature of wind power severely affects the stability of the electric grid. It is well known that power quality and project economics are adversely affected by increased intermittency and variability. Thus, an accurate, simple, and powerful metric for comparing the variability of different power sources would serve as a useful figure of merit. The current coefficient of variation metric has significant shortcomings for quantifying the variability of a stochastic power source's output. This paper exposed those shortcomings in detail and a new metric is proposed for quantifying the variability of power from a wind farm in particular or for an intermittent power source in general. The proposed metric is calculated from the spectral decomposition of a power source's autocorrelation function. The intermittency metric compares the components of the wind power that vary with infra-diurnal frequencies to the mean power output.