Experimental study on mitigating vortex-induced vibration of a bridge by using passive vortex generators

Highlights

• We proposed a 3D spanwise-varying control, passive vortex generators (PVG), to mitigate VIV of bridges.

• The PVGs perform well on suppressing VIV with and without railings through wind tunnel tests.

• The PVGs which generate stramwise vortex pairs suppress the spanwise vortices shedding effectively.

Abstract

It has been reported that three-dimensional spanwise-varying control is very efficient in suppressing vortex shedding behind bluff bodies. Therefore, a new passive control method, spanwise-varying passive vortex generator (PVG), was proposed to mitigate the vortex-induced-vibration (VIV) of a bridge. The PVGs can directly generate streamwise vortices, which behave like secondary wake instability modes (3D instability modes). Therefore, the PVGs can cause 3D instability of spanwise vortex shedding and suppress VIV. In this study, isolated PVGs were arranged on the lower surface of the bridge in spanwise direction, and their control effects on mitigating VIV were investigated through wind tunnel tests. The results show that as PVGs are arranged with spanwise distance of 1–3H (H is the height of the bridge) and the PVG height is 0.1–0.25H, both vertical and torsional VIV can be completely suppressed, and the trailing edge is the key region to suppress VIV. Further, the control effects of PVG are barely influenced by the boundary layer of the bridge. The spectrum analysis of the wake velocity shows that the PVG can completely suppress the vortex shedding near a wake, and thereby resulting in the vanishing of VIV.

Introduction

As wind passes around the long span bridge deck, the spanwise vortices will shed in the near wake. If the vortex shedding frequency is close to the natural frequency of the bridge deck, it can cause vortex-induced resonance. A vortex-induced vibration (VIV) is one of the major issues in long-span bridge vibration. Although the VIV is a self-limited vibration, it can introduce large displacements and cause discomfort to the drivers (Battista and Pfeil, 2000, Frandsen, 2001, Fujino and Yoshitaka, 2002, Larsen et al., 2000). In addition, vortex-induced resonance always occurs at low wind speeds and can lead to long-term fatigue damage.

Many types of control methods for suppressing VIV have been proposed. However, passive flow control methods are usually selected because of their reliability and simplicity. Fujino and Siringoringo (2013) provided a review of these control methods in bridge engineering. Conventional passive flow control attachments, which are always attached to the leading and trailing edges, include guide vanes, fairing, flaps, wind spoiler and deflector (Fujino and Siringoringo, 2013). Among them, guide vanes are more popularly used in suppressing VIV since they provide better control effect. Larsen et al. (2000) installed guide vanes on the Greatbelt bridge deck, which was then subjected to severe VIV, and observed a reduction in the amplitude of the VIV from the wind tunnel test results. Later, Larsen et al. (2008) investigated the vortex response of a twin box bridge with and without the guide vanes at three different Reynolds numbers and observed that the guide vanes had no obvious control effects on the VIV at low Reynolds number as no sufficient flow rate passed it with high boundary layer. Therefore, it indicated that the control effects of the guide vanes were sensitive to the Reynolds numbers. Further, the aerodynamic countermeasures above always increase the maintenance efforts and construction costs since these attachments need be installed along with the bridge deck. In addition, Larsen and Wall (2012) explored the effect of the shape on the VIV of trapezoidal box girder bridge decks and found that it was possible to derive a virtually vibration-free deck. However, this approach could fail if the shape was restricted because of the other requirements. Additionally, the shape of a vibration-free deck might also not be considered esthetic.

As the regular vortex shedding is the main source of the VIV, suppressing spanwise vortex shedding is an efficient way to mitigate the VIV of a bridge deck. The three-dimensional spanwise-varying control, which directly or indirectly changes the spanwise aerodynamic shape of a bluff body, is more efficient than the spanwise homogeneous method (based on 2D framework) on attenuating vortex shedding. Kim and Choi (2005) compared the momentum required for drag reduction of a 3D spanwise-varying control method and a spanwise homogeneous method (base bleeding). The results indicated that the momentum required of the three-dimensional forcing was significantly less than that of a conventional two-dimensional forcing (base bleeding). The underlying mechanism of the three-dimensional spanwise-varying control can be ascribed to the action of the secondary wake instability, which is also called the 3D instability of spanwise vortices (Hwang et al., 2013). There are mainly two types of secondary instabilities existing in a bluff body wake, namely the Modes-A and Modes-B (Williamson, 1996a, Williamson, 1996b, Wu et al., 1996). Both instability modes act as waves in the spanwise wake vortices, which evolve into pairs of counter-rotating streamwise vortices connecting the Karman vortices. These secondary streamwise vortices cause the spanwise dislocation of the Karman vortices and the energy transfer from the spanwise vortices to the streamwise vortices (Doddipatla et al., 2008). Because of the highly efficient control effects of the 3D spanwise-varying control, its application for mitigating vortex shedding and/or achieving drag reduction has attracted much attention. The numerical simulation results of Darekar and Sherwin (2001) showed that for providing spanwise sinusoidal perturbations at the leading and trailing edges for a square cylinder, the spanwise vortex shedding was highly suppressed at 10 < Re(d) < 150 (d being the character length of the bluff body) when the perturbation wavelength was 5.6d. Kim and Choi (2005) investigated the control effects of spanwise-periodic blowing/suction on reducing mean drag of a circular cylinder at the Reynolds numbers between 40 and 3900. The numerical results showed that the spanwise-periodic blowing/suction attenuated or annihilated the Kármán vortex shedding and thus significantly reduced the mean drag and the drag and lift fluctuations. Dobre et al. (2006) significantly attenuated the spanwise vortex shedding around a square cylinder by using spanwise sinusoidal perturbations with wavelengths of 2.4 d at the trailing edge at Re(d) = 12 500. Park et al. (2006) achieved 33% base pressure recovery at the Reynolds numbers up to Re(d) = 60 000 using spanwise arrays of vertical rectangular tabs at the trailing edge. Deshpande and Sharma (2012) used segmented trailing edges to achieve up to 34% base pressure recovery and nearly completed the suppression of vortex shedding at Re(d) = 8750. More related studies can be found in a review by Choi et al. (2008).

To the authors’ knowledge, very few studies have focused on the 3D spanwise-varying control effects on suppressing the VIV of bridges. El-Gammal et al. (2007) successfully suppressed the VIV of a girder by using spanwise periodic perturbation method (SPPM). In their study, the leading and the trailing edges of the deck were modified as a wave in the spanwise direction. However, the shape of the girder was changed dramatically, which could cause esthetic problems and affect other performance aspects. In this paper, the PVGs based on the 3D spanwise-varying control are proposed to attenuate the spanwise vortex shedding and mitigate the VIV of a bridge deck. The PVGs have been widely applied in airfoils to delay the boundary layer separation (Lin, 2002). The streamwise counter-rotating vortex structures (see Fig. 1, Von Stillfried et al., 2010) induced by vortex generators force the higher momentum fluid from the free stream towards the wall and increase the near-wall momentum. Consequently, the stabilised flow leads to the delay or prevention of separation of the boundary layers. As stated above, the streamwise vortices are able to cause the spanwise dislocation of the Karman vortices or even destroy the spanwise vortices. Therefore, the PVGs can be used to generate the streamwise counter-rotating vortex pairs directly and to activate the corresponding mode of secondary instability to suppress the spanwise vortex shedding and mitigate the VIV. Although the PVGs are widely used in boundary layer separation control, their application in controlling the VIV of bluff bodies is missing. Therefore, the main objectives of this study are to propose and design the spanwise-varying PVGs control method and then to investigate its control effects on suppressing the VIV of a bridge deck by wind tunnel tests.