Wind-induced Internal Resonance Of Cable-hanger System On Suspension Bridge

Wind-induced oscillations of hangers on a domestic super-long-span suspension bridge are investigated based on field observations. Wake galloping, vortex-induced resonance, and wind-rain-induced vibration are excluded by qualitative analysis. Theoretical analysis, numerical simulation and simplified wind tunnel test demonstrate successively that it's a kind of hanger resonance induced by the buffeting of the main cables. Due to the dense natural frequencies of hangers, the broad-spectrum feature of structural buffeting, and the wind-induced stochastic cable responses abundant in modal components, the resonance of hangers occurs when the main cable's vibrational frequencies of adequate buffeting energy are sufficiently close to those of the hangers. Hence, it's not the hangers themselves but the main cables that are responsible for the energy absorbing from the turbulence. Compared to the modal mass of a main cable, the mass of a single hanger is too small to pose substantial effects to the oscillation of the main cables, and therefore a steady supply of energy from the main cable to the hangers can be formed. Analytical results indicate that the amplitude of the hangers could be sensitive to its natural frequency, wind speed, and the damping ratios. On the other hand, aerodynamic interferences among the 4 separated branches of each hanger could further complicate the motions of the hangers, resulting sometimes in un-synchronous oscillations of the 4 separated branches. Furthermore, MTMD mounted on the main cables could suppress effectively the cable motions close to the natural frequencies of hangers, which in turn results in significant reductions of the vibration of the hangers. The major work and conclusions are as follows:(1) First, the effects of wind loads on bridge, the types of cable oscillation as well as corresponding control methods are presented. The porperties of the observed hanger oscillationsin field are compared with the known types of cable vibration. The wake galloping, vortex-induced resonances, rain-wind induced vibration, and bridge tower wake induced vibration are excluded in terms of qualitative analysis. The finite element model of the bridge has been built with element refinement of hangers, and the dynamic properties of the full structure are analyzed. The natural modes of the hangers are extracted. It is obtained that the natural frequencies of hangers are low and closely distributed, and the natural modes of the main cables and hangers interweave in very small frequency intervals, indicating the possibility of internal resonance in the cable-hanger system.(2) The stochastic wind field acting on the girder and main cables is simulated by the method of harmonic synthesis. The buffeting wind loads on the bridge are presented. The wind load fluctuations are exerted on the bridge deck and the main cables. The wind loads on the hangers are neglected to avoid confusion in qualitative analysis. The buffeting histories of the full bridge are numerically simulated. The lateral displacement time histories of the hangers are extracted at the hanger center, the upper end on the main cables and the lower end on the bridge deck, based on which the spectrum analysis is performed. Numerical results indicate it is the main cables rather than the hangers that trap energy from the oncoming turbulent wind field, and then, energy passes from the main cable to the hangers. The lateral dynamic shapes of a long hanger and the main cable to which it is attached are plotted, which illustrated that the resonances of the hangers are induced by higher-ordertiny(not perceptible) motions of the main cables. The resonance could be sensitive to positions along the bridge deck(the natural frequencies of hangers itself: length and material characteristics et al), wind speeds and damping ratios. The incomplete synchronous oscillation of the 4 steel ropes of each hanger can be ascribed to the aerodynamic interferences among them, including both from the quasi-steady shear flow and from the unsteady signature turbulence(or regular vortex shedding).(3) A simplified experimental model of the main cable-hanger system is built, and the damping ratios of the main cable and the hanger are tested separately by mean of self-vibration. Resorting to free beat vibration, the natural frequency of the calbe-hanger system is adjusted to the target frequency. The wind tunnel test is performed in grid-induced fluctuating wind fields. The hanger can been closed in a glass shield, which could reflect the influences of the wind loads exerted on the hanger itself by comparing the two cases with and without the glass shield. The experimental results indicate that, disregarding whether or not the hanger is enclosed, the internal resonance of cable-hanger system exists at all tested wind velocities. Therefore the kind of wind-induced resonanceis verified.(4) In virtue of the mechanism, a new method is put forward aiming at the mitigation of the hanger resonance. The method involves a multiple tuned mass damper(MTMD) mounted on the main cables, which could suppress the motion components of the cables at the natural frequencies of the hangers, and finally result in the suppression of the hanger resonances. Influences of mass ratio, damping ratio, frequency ratio, the location of installation, and quantity of TMD are investigated. The effect of mitigation is compared with that when dampers are employed. The results of numerical calculation indicate that the MTMD method has a significant suppression effect on the hanger resonances. Moreover, in contrast to the poor effect of dampers, the effect of MTMD is insensitive to the structural damping and wind speeds.