Wind speed prediction for wind farm applications by Extreme Value Theory and Copulas

• We advance a Generalized Pareto distribution for the modeling of the probability distribution function's tail.
• We model the wind turbine failures through a Poisson distribution.
• We use a Copula to consider the correlation between wind turbines that compose the wind farm.
• We compare the energy evaluated through the proposed procedure and a simple Weibull based model.

In this paper we use risk management techniques to evaluate the effects of some risk factors that affect the energy production of a wind farm. We focus our attention on three major risks: wind speed variability, wind turbine failures and correlations between produced energy.As a first contribution, we show that the Weibull distribution, commonly used to fit recorded wind speed data, underestimates rare events. Therefore, in order to achieve a better estimation of the tail of the wind speed distribution, we advance a Generalized Pareto distribution. We considered one aspect of the wind turbines reliability by modeling their failure events as a compound Poisson process. Finally, the use of Copula enables us to consider the correlation between wind turbines that compose the wind farm. Once this procedure is set up, we show a sensitivity analysis and we also compare the results from the proposed procedure with a simplistic energy prediction using the Weibull distribution.