Digital generation of surface-pressure fluctuations with spiky features

The paper presents a new method for generating artificial surface-pressure time series focused on the simulation of non-Gaussian spiky fluctuation features. The method is based on the Fourier representation of time series with special selection of random Fourier phases. The unique properties of the Fourier phases are utilized for generation and control of the large-amplitude spikes which are the essential feature of surface-pressure signals typically observed around the corners and edges of a building. The synthetic time-series generation, focused on extreme-value distribution properties, are presented for a given target surface-pressure signal with skewed spiky features and non-Gaussian probability distribution. (C) 1998 Elsevier Science B.V. All rights reserved.