A review of time series models for the long term modeling of wind speed in distribution network planning

In this work, a review of univariate time series mo dels for the long term modeling of wind speed is perform ed. These models intend to help the distribution system operators to account for the stochastic nature of w ind during the network planning phase. Two groups of criteria are proposed for comparing the performance of the models. The first one is focused on the representativeness (i.e. the ability of the model t o reproduce the statistical properties of the real da ta), whereas the second one is linked to computational issues, such as the size of the historical dataset which is needed to obtain a given level of representativenes s, or the computational burden associated to the training