A model of probability density function of non-Gaussian wind pressure with multiple samples

• The monotonic limits of Hermite moment models were described numerically by skewness and kurtosis.
• The probability distribution of non-Gaussian peak factor was formulated in terms of shape parameters of Hermite models and Gaussian peak factor.
• The probability distribution of extreme wind pressure coefficients was formulated in terms of the mean, standard deviation and non-Gaussian peak factor.
• The extreme wind pressure coefficient was estimated and compared.

The extreme value of non-Gaussian wind pressure coefficients is usually estimated by fitting the probability density function (PDF) of maximum or minimum values while a large number of observations except the peak values in the measured samples are discarded. The implicit or explicit translation between non-Gaussian and Gaussian histories can also be utilized to estimate the extreme value of non-Gaussian wind pressure coefficients while the randomness of aerodynamic effect is not taken into account. The Hermite moment models are summarized and applied to formulate the PDF of non-Gaussian peak factors which is expressed as the joint PDF (JPDF) of the shape parameters of the Hermite moment model and the PDF of the translated Gaussian peak factors. After the variations of mean, standard deviation and non-Gaussian peak factor are considered, an innovative and analytical PDF formula of extreme wind pressure coefficients for multiple samples, which is expressed as a function of the JPDF of mean, standard deviation and the PDF of non-Gaussian peak factor, is presented in this paper. The theoretical developments are applied to establish the PDF and cumulative density function (CDF) of the negative peak wind pressure coefficients with multiple samples. It is verified that the analytical probability distribution model is a reasonable model to estimate the peak pressure coefficients.