Simulation of Wind Scenarios with R-vine Copulas under Hydrological and Climate Dependence

The growing demand for sustainable energy has amplified the importance of integrating renewable sources. This study proposes a methodology based on R-vine copulas for simulation of wind generation scenarios, incorporating the Influent Natural Energy (ENA), its time lags and the Oceanic Niño Index (ONI). ONI was empirically modeled from sea surface temperature anomalies in the Niño 3.4 region, using ERSSTv5 data from 1980 to 2023. The R-vine structures were selected monthly using the Bayesian Information Criterion (BIC) and estimated by maximum likelihood. The methodology allowed the generation of 2,000 monthly scenarios of the wind capacity factor (FC), preserving nonlinear dependencies, temporal structure and behavior in tails. The results showed the ability of the approach to adequately represent the joint dynamics between hydrological, climatic and wind variables, contributing to applications in energy planning under climate uncertainty.