Theoretically estimated peak wind loads

Abstract

Current estimation of peak pressure coefficients and peak wind loads on structures in the ASCE 7 Standard [American Society of Civil Engineers, ASCE Standard, Minimum Design Loads for Buildings and Structures, ASCE 7-02, ASCE, Reston, VA, USA, 2002] is based on the assumption that they are distributed normally. However, this assumption is erroneous in the case of low-rise structures because the time varying pressures and loads along roof edges and ridges have been observed to be generally non-Gaussian [H.W. Tieleman, Z. Ge, M.R. Hajj, T.A. Reinhold, Pressures on a surface-mounted rectangular prism under varying incident turbulence, J. Wind Eng. Ind. Aerodyn. 91 (2003) 1095–1115]. In this article, a new procedure is used to evaluate from one individual non-Gaussian sample record statistics of peak pressure or peak load coefficients. The initial step for the analysis requires the identification and evaluation of the appropriate marginal probability distribution. The results reveal that the distribution of the time histories of surface pressure and load coefficients is well represented by the gamma distribution, whose parameters can be adequately evaluated from the theoretical moment estimators. The corresponding distribution of the peaks of the sample records that represents either the pressure coefficients or the load coefficients, can then be obtained using a standard translation process approach. This information yields the mean and the standard deviation of the sample peaks, which are then used to determine the extreme coefficients associated at any selected probability level of non-exceedence. This latter step can be made by assuming that the distribution of the peaks follows the Extreme Value Type I (Gumbel) distribution. The analysis is applied to pressure measurements on the 1:50 scale model of the experimental building at the Wind Engineering Research Field Laboratory (WERFL), and executed in the Clemson boundary-layer wind tunnel over a range of incident turbulence intensities.