On the Capacity Factor of Distributed Wind Generation in Droop-Regulated Microgrids

This paper presents the methodological basis for the computation of the capacity factor (CF) of the power delivered by an off-grid droop-regulated microgrid when the power is injected by distributed stochastic generation. More particularly, the paper is focused on wind generation and its modeling particularities. It shows how the availability of power can be efficiently sampled by a low variation Latin Hypercube Sampling (LHS) method, supplemented with a restricted pairing technique to account for correlation among generating units. The paper also presents a minimization problem that, combined with a Fischer-Burmeister-based formulation, permits obtaining the maximum deliverable power at each sample. This paper additionally shows a first numerical analysis of a 14-node microgrid supplied by five Enercon E40. The analysis includes an investigation into the best model that represents the CF of the microgrid, and highlights the effects of correlation and maximum loadability.