Probabilistic modeling of wind pressure on low-rise buildings

A third order Hermite polynomial PDF model is adapted from the literature and investigated as a robust model to represent fluctuating pressures over the surface of low-rise buildings. The relationship between skewness, kurtosis and the Hermite model shape parameters is numerically inverted and an approximate solution is developed to improve the accuracy and simplify the application of the Hermite model to measured data. The effective region with respect to skewness and kurtosis is established for the Hermite model. It is shown that wind pressure data from full scale and wind tunnel model low rise buildings fall within or very close to this effective region. Common PDF models utilized in wind engineering literature are compared with the Hermite PDF model for mildly and strongly non-Gaussian wind tunnel and full scale pressure datasets. The superior performance of the proposed Hermite PDF model is demonstrated.

► A Hermite PDF model is shown to be robust and accurate for pressures over a low-rise building.
► An expression relating the model shape parameters and the higher moments is developed.
► The admissible region for the PDF model is presented as a function of skewness and kurtosis.
► The superior performance of the Hermite PDF model is demonstrated against common PDF models.
► A goodness of fit test is developed for the tail of the distribution.