Physical simulations on wind loading characteristics of streamlined bridge decks under tornado-like vortices.

Tornado risks for highly important long-span bridges in tornado-prone regions can not be neglected. Rigid-model wind pressure measurements on a streamlined bridge deck were conducted using a tornado vortex simulator, to clarify tornado-induced surface pressure distributions, aerodynamic load coefficients and total force coefficients. We focused on two main parameters: swirl ratio and horizontal distance from tornado center to deck. Obvious discrepancies were observed between tornado-induced wind loading and results from conventional boundary-layer wind tunnels. Strip theory was not applicable since pressure distributions vary for different sections of the deck along the bridge axis. The absolute values of mean pressure coefficients for the deck section near tornado center are largest among all tested sections along the bridge axis. The most unfavorable mean sectional drag force coefficients were found when the bridge model is located at the tornado core radius and largest mean sectional lift force coefficients at the tornado center. With increase in swirl ratio, magnitudes of mean pressure coefficients as well as mean and fluctuating sectional aerodynamic load coefficients become smaller. The unfavorable locations for fluctuating rolling moment coefficients are different from those for mean values, which indicates that the quasi-steady assumption becomes invalid for tornado-induced rolling moment coefficients. Total force coefficients over the entire deck were investigated to evaluate the non-uniform loading characteristics. The findings will be helpful for predicting tornado-induced responses and risks for highly important long-span bridges. • Tornado-induced wind loads on a streamlined bridge deck was carried out. • The force coefficients are position-dependent and strip theory is not applicable. • Largest drag coefficients are at tornado radius and lift force at tornado center. • Magnitudes of pressure and force coefficients decrease with increase in swirl ratio. • Results are helpful for predicting tornado-induced responses and risks for bridges. [ABSTRACT FROM AUTHOR]