Probabilistic models for 3D structural demand considering the ground motion group size and the angle of seismic incidence

Abstract

The probabilistic seismic demand of selected engineering demand parameters obtained from three-dimensional analysis of RC buildings is analysed by examining their statistical distribution considering multiple angles of seismic incidence and various ground motion group sizes. Six probability distribution models are fitted to the engineering demand parameter datasets and their performance is examined based on quantile-quantile plots as well as on the corresponding coefficient of determination and the root mean square error. The results show that using more than one angle of seismic incidence does not affect the type of demand distribution that best fits the datasets in the majority of the cases that were analysed. The probability distribution models that present a higher performance are further examined using goodness-of-fit tests with power against specific distribution models. The superiority of the fitting performance of the lognormal and the Gamma distributions is demonstrated, while the particular effect of the number of angles of seismic incidence on the empirical data is discussed along with suggestions regarding the distribution selection in case a higher number of angles is used.