Stochastic analysis and simulation of wind speed and solar radiance for sustainable energy production

This thesis focuses on the stochastic analysis and simulation of meteorological wind and solar radiation daily data, collected from Schiphol Airport in Amsterdam. The primary objective is to generate synthetic time series that preserve key statistical and dependency characteristics of the original data, for applications in sustainable energy production. The analysis includes the study of seasonality, autocorrelation, cross-correlation, and memory persistence structure, based on the Hurst–Kolmogorov dynamics. Synthetic data are produced using the Symmetric Moving Average (SMA) model, which is evaluated for its ability to preserve essential stochastic features. Additionally, reordering methods are applied to generate non-Gaussian time series with specified dependence structures and preserved inter-variable correlations. This project investigates furthermore, distributional biases introduced during simulation, methods for maintaining the seasonality of the skewness coefficient, and the preservation of cross-correlations between physical processes governed by different distributions. The results contribute to a deeper understanding of the stochastic behavior of natural phenomena and enhance modeling capabilities for applications in renewable energy systems.