Quantitative characterization of uncertainty levels of intermittent power sources

Abstract

This paper designs a statistical quantification towards the intermittent power uncertainty in power systems. A negative-exponential forecast uncertainty function is constructed to represent the relationship between the statistics of forecast error of a single intermittent power source and time advance. Subsequently, other kinds of statistical functions are proposed to characterize the statistical uncertainty of multiple intermittent power sources and all power sources, namely the sum statistical functions, the equivalent statistical functions, and the contour statistical functions. Based on a large amount of historical observations, these functions are employed to statistically quantify the forecast uncertainty of a single intermittent power source, multiple intermittent power sources as well as all power sources. Historical data sampled from real wind farms and solar sites demonstrates the effectiveness of the proposed method.

Selection of a Critical Time Scale of Real-Time Dispatching for Power Systems With High Proportion Renewable Power Sources

“…Here, this paper use the uncertainty function to describe the relationship between the power prediction error and the prediction lead time of the URES, previously defined in [35], as follows: (8) where, A ures is the amplitude of the uncertainty function, and τ ures is the time constant of the uncertainty function. E ures (t) is the power prediction error of URESs affected by the prediction lead time.…”