A heuristic for the synthesis of credible operating states in the presence of renewable energy sources

Abstract

Experience has shown the limitations of deterministic criteria when accommodating the intrinsic uncertainties associated to modern power systems. Hereof, probabilistic risk assessment represent a powerful enhancement in order to ensure the overall power system reliability rather than a worst-case scenario analysis. This paper presents a general-purpose methodology intended to generate plausible operating states. The main focus lies on the generation of correlated random samples using a heuristic of the NORmal-to-Anything (NORTA) method. The proposed methodology was applied to model wind generation in the Danish Western power system, analyzing the effect of the marginal distributions and errors in the correlation matrix definition.