I. Deligiannis, Investigation of double cyclostationarity of wind for application to stochastic forecast models, Diploma thesis, 119 pages, Department of Water Resources and Environmental Engineering – National Technical University of Athens, July 2016.
Τhe wind speed’s behavior is examined in an hourly, monthly and over-annual scale. First of all, 1,800 global wind stations are selected to be investigated, in order to ensure that a double periodicity exists (daily and annual cycles). The double cyclostationarity of wind speed is justified through the correlation among wind speed, temperature and radiation. In addition, a deterministic model is selected among several ones and applied to the wind speed and temperature data. What is more, the efficiency of some reanalysis data is checked by observing if they describe this double periodicity. Finally, a stochastic investigation of wind speed is conducted, by studying the long term persistence of annual mean wind speed in 1300 stations globally.