Data-driven reconstruction of wind speed randomness in an urban area

Abstract

Abstract This study proposes two diffusion models to analyze the wind speed variability in an urban area. The analysis is based on annual time series data collected from fourteen weather stations. A basic criterion has been suggested to categorize these stations based on the variance of the stochastic process for the stationary case. This criterion can be used in studies of air pollution, wind energy, and other related fields where the geographical classification of weather stations is not feasible. The Kramers-Moyal (KM) coefficients and kernel-based regression (KBR) have been utilized to estimate the drift and diffusion terms. The numerical solution of the proposed Langevin equation was used to calculate the statistical properties of the process, taking into account the variance values for station classification. The results show that only two Langevin models are required instead of the original fourteen, based on the variance values. This demonstrates that it is feasible to establish models using basic statistical properties of time series when geographical classification is not possible.