Civil and Environmental Engineering Department Seminar: Wind Effects on Flexible Structures: A New Perspective

Seminar | February 27 | 10-11 a.m. | 542 Davis Hall

 Teng Wu

 Civil and Environmental Engineering (CEE)

Wind effects on flexible structures such as high-rise buildings and long-span bridges, governed by the Navier-Stokes equations, are not adequately represented by a conventional linear analysis framework. This shortcoming is becoming significant for contemporary structures, as their increasing heights/spans and constantly changing cross-sections make them more sensitive to nonlinear and unsteady aerodynamic load effects. This presentation focuses on developing a unified analysis framework for nonlinear aerodynamics in the context of cable-supported bridges by responding to the following key questions: (1) What are the typical nonlinear behaviors observed from wind-tunnel studies and full-scale observations? (2) What are the effects of nonlinearity and unsteadiness on bridge aerodynamics? (3) What is the ability of existing nonlinear models to capture nonlinear and unsteady effects? (4) Is it possible to go beyond the current nonlinear models, and establish more effective nonlinear unsteady low-dimensional modeling techniques? To this end, the higher-order spectral scheme is utilized to identify aerodynamic nonlinearity. The effects of nonlinearity and unsteadiness on bridge aerodynamics are evaluated by comparing aerodynamic responses derived from various semi-empirical models. Current models set in the conventional analysis framework are reviewed to understand their ability in simulating nonlinear unsteady aerodynamics. Several advanced low-dimensional modeling techniques, characterized by different levels of analysis of nonlinearity and unsteadiness, are then proposed. These include an artificial-neural-network approach, a nonlinear moving-average model, and a Volterra series-based model. The fidelity with which the proposed approaches are able to simulate nonlinear bridge aerodynamics is verified through data based on wind-tunnel tests or computational fluid dynamics.

 leorig@berkeley.edu