Synthesis of non-Gaussian wind pressure time series on low building roofs

Abstract

This paper presents an efficient and practical method for the digital generation of univariate non-Gaussian wind pressure time series on low building roofs. The method, based on the Fast Fourier Transform (FFT) approach, essentially inverts the Fourier coefficients which are a linear combination of Fourier amplitude and phase. In this study, the Fourier amplitude part is assumed to be known. The Fourier phase capable of inducing non-normality to the time series is carefully modelled and a simple stochastic model with a single parameter is suggested for its simulation. The computation of this single parameter is accomplished by minimizing the sum of the squared errors in higher order statistics such as skewness and kurtosis. The simplicity and effectiveness of this methodology have been demonstrated using several measured non-Gaussian pressure data from various low building roofs under different conditions.