In wind tunnel tests of bridges for buffeting arising from fluctuating wind
velocity, one of the items requiring attention is how to simulate the natu
ral wind characteristics of the construction region. In wind tunnel tests o
f long-span cable-stayed bridges under construction, when two types of turb
ulent boundary layer with different wind characteristics were used, there w
as a tendency for the match between the wind tunnel test values and buffeti
ng analysis values to vary with the type of turbulent boundary layer. Horiz
ontal buffeting was considered for which there are fewer previous examples
than for Vertical buffeting. As a result of pursuing the cause of the lack
of matching, it was found that the aerodynamic admittance used in the analy
sis and the measured aerodynamic admittance were quite different. This diff
erence arose from the fact that the correlation of the fluctuating velocity
spanwise was smaller than the value assumed in the analysis, Also, accordi
ng to past research, there are many examples in which strip theory accordin
g to which the fluctuating wind velocity correlation and the fluctuating ae
rodynamic correlation are equivalent is not valid. Here, it is also conside
red whether or not the strip theory is valid for horizontal buffeting. (C)
1999 Elsevier Science Ltd. All rights reserved.