Stochastic analysis and simulation of hydrometeorological processes associated with wind and solar energy

We investigate the existence of long-term persistence by estimating the Hurst coefficient for wind speed and sunshine duration variable on an annual scale in Europe. The data we used were retrieved in the form of daily time series, from the KNMI, ECA & D and METEIREANN databases, and refer to a period of at least 70 years. Wind speed records collected from 20 measuring stations were treated appropriately according to their transformation to form a single measurement altitude; as well as sunshine duration records from 21 measuring stations. Hurst coefficient was estimated using the method of linear regression and the method of least squares based on the standard deviation (LSSD); the existence of persistence was observed for both variables. Moreover, we found characteristic values of the Hurst coefficient for both variables, around which the coefficients of the measuring stations we surveyed are gathered. Furthermore a simultaneous generation of synthetic time series of these two variables on a yearly, monthly and daily scale is attempted using a multivariate stochastic model through the computing system Castalia. Two applications of the program are made, one for the raw data and one using a logarithmic transformation of the sunshine duration variable before entering into the program because of its blocked definition field. We conclude that using the logarithmic transformation, Castalia is able to generate satisfactorily synthetic time series of these two variables.