Slender structures would experience excessive levels of vibration under the action of wind. Galloping is characterized as a periodical, self-excited, large-amplitude oscillation, which can lead collaspse of structures at a very short time period. Galloping response is conventionally predicted by using the classic quasi-steady theory. However, the quasi-steady theory fails to predict low wind speed galloping instabilities of structures and is not suitable to predict galloping instabilities of inclined prisms as the effect of structural oscillation (unsteady effect) is excluded in the quasi-steady theory. This thesis aims to understand unsteady crosswind forces acting on slender vertical and inclind structures and to address the shortcomings of the quasi-steady theory in predicti...[ Read more ]

Slender structures would experience excessive levels of vibration under the action of wind. Galloping is characterized as a periodical, self-excited, large-amplitude oscillation, which can lead collaspse of structures at a very short time period. Galloping response is conventionally predicted by using the classic quasi-steady theory. However, the quasi-steady theory fails to predict low wind speed galloping instabilities of structures and is not suitable to predict galloping instabilities of inclined prisms as the effect of structural oscillation (unsteady effect) is excluded in the quasi-steady theory. This thesis aims to understand unsteady crosswind forces acting on slender vertical and inclind structures and to address the shortcomings of the quasi-steady theory in predicting galloping responses of the prisms.

A new wind tunnel test rig was devised to perform a forced vibration wind tunnel test and a hrybrid aeroelastic-pressure wind tunnel test to achieve the objectives. In the wind tunnel tests, unsteady crosswind forces of vertical and inclined prisms were recorded during forced and aeroelastic oscillations. The effect of structueal oscillation on the unsteady crosswind forces was therefore included.

The hybrid aeroelastic-pressure test (HAPT) system was a weakly nonlinear system and under wind-off conditions, non-wind-induced aerodynamic forces that were induced by the interaction between an oscillating test model and its surrounding air were considerable. Physical nonlinearities (nonlinear damping and stiffness) of the test system were identified using a modified wavelet transform method. Non-wind-induced nonlinearities (wind-off conditions) of the system were determined using the forced vibration test technique. The results indicated that both the nonlinear damping and stiffness increase with the amplitude of oscillation. The non-wind-induced nonlinearities were considerable and cannot be neglected in predicting galloping responses.

From the forced vibration test, the characteristics of unsteady crosswind forces of vertical and inclined prisms were analyzed in light of spectra of generalized unsteady crosswind forces, spectra of pointwise pressures, and Strouhal numbers. The aerodynamic damping of the prisms was identified based on unsteady crosswind forces. Nonlinear mathematical models for the aerodynamic damping were developed and were subsequently used for predicting self-excited oscillations of the prisms. The results indicated that both the amplitude of vibration and forward or backward inclination have significant effects on the aerodynamics of vertical and inclined prisms, especially in the range of wind velocities that the vortex lock-in occurs. The nonlinear mathematical models for the aerodynamic damping can give better predictions in galloping responses of the vertical and inclined prisms than the classic quasi-steady theory. However, considerable differences exist between the model calculation and the experimental result in estimating critical galloping wind speeds. This is probably because the mathematical models were developed based on the result of the forced vibration wind tunnel test, in which the test model cannot experience the same force as a structure in field conditions.

To precisely predict galloping instabilities of vertical and inclined prisms, unsteady galloping forces were obtained from the aforementioned hybrid aeroelastic-pressure test (HAPT). They are determined by deducting the non-wind-induced aerodynamic forces from unsteady crosswind forces directly observed from the test. Galloping responses were therefore calculated by the unsteady galloping force and they were compared with quasi-static calculations and experimental results. It was found that the quasi-steady theoy is not suitable for predicting galloping instabilities of slender prisms whereas unsteady galloping forces can accurately predict galloping instabilities. Nonlinear mathematical models that are functions of aerodynamic damping and stiffness were developed based on an equivalent energy principle to predict galloping instabilities of vertical and inclined prisms. The results showed that a 3rd-order nonlinear mathematical model can precisely predict galloping instabilities of prisms which have different mass-damping ratios. Furthermore, both time history responses and root-mean-square (RMS) tip responses of the vertical and inclined prisms were calculated by the nonlinear mathematical models. These results were in close agreement with experimental results, indicating that the developed nonlinear mathematical models can well address the shortcomings of the classic quasi-steady theory in predicting galloping responses of slender prisms.