Estimation of Extreme Wind Load on Structures and Claddings

Abstract

A modified Cook-Mayne method is proposed for extreme wind load estimation. A technique that can simulate true r-order extreme values of a correlated random process is adopted in this method to estimate the annual multiorder maximum 10-min mean wind speeds. These estimated extreme wind speeds are then used for extreme wind load estimation on structures. Results show that the autocorrelation of the macrometeorological fluctuating wind can have significant effects on the estimated extreme wind load. A big difference is found when comparing the design fractile of the extreme wind load coefficients obtained from the modified method by considering the autocorrelation of wind speeds and the existing full-order method in which only independent wind speeds are adopted. The order of annual maximum wind speeds required for wind load estimation is also examined. The number of orders that is needed for 50-year return period wind load design varies greatly—from 2 to 50—and depends on the application conditions. The design fractile of extreme wind load coefficient distribution for a 50-year return period can be increased to 93% from the 79% recommended in the existing method.