Quantifying short-term wind power variability

Quantifying wind power generation variability is essential to power system operators, since wind variability has a significant effect on the system operating reserve requirements. An underestimation of these requirements can severely affect the reliability of the system. Moreover, quantifying wind power generation variability, regardless of its uncertainty, can provide important characteristics a generator must have so as to accommodate wind fluctuations. The purpose of this paper is to provide a new metric to characterize short-term variability of wind power generation or any other source. The new metric estimates the size of the interval within which the variable source's output values lie over a desired time frame, irrespective of the output's forecasting error levels. A case study exposes the shortcomings of the prevalent variability metric used in power systems, which is based on the spread of the variable source's output step-changes over various time frames. The comparison of the two metrics using real wind generation and demand data reveals that the proposed metric can effectively capture wind variability, and is capable of accommodating demand variability for up to 20.87% more instances.