Vortex-induced vibrations of two side-by-side circular cylinders with two degrees of freedom in laminar cross-flow

2019 ◽  
Vol 193 ◽  
pp. 104288 ◽  
Author(s):  
Weilin Chen ◽  
Chunning Ji ◽  
Dong Xu
2012 ◽  
Vol 204-208 ◽  
pp. 4598-4601
Author(s):  
Jie Li Fan ◽  
Wei Ping Huang

The two-degrees-of-freedom of vortex-induced vibration of circular cylinders is numerically simulated with the software ANSYS/CFX. The VIV characteristic, in the two different conditions (A/D=0.07 and A/D=1.0), is analyzed. When A/D is around 0.07, the amplitude ratio of the cylinder’s VIV between in-line and cross-flow direction in the lock-in is lower than that in the lock-out. The in-line frequency is twice of that in cross-flow direction in the lock-out, but in the lock-in, it is the same as that in cross-flow direction and the same as that of lift force. When A/D is around 1.0, the amplitude ratio of the VIV between in-line and cross-flow in the lock-in is obviously larger than that in the lock-out. Both in the lock-in and in the lock-out, the in-line frequency is twice of that in cross-flow direction.


Author(s):  
Andrew E. Potts ◽  
Douglas A. Potts ◽  
Hayden Marcollo ◽  
Kanishka Jayasinghe

The prediction of Vortex-Induced Vibration (VIV) of cylinders under fluid flow conditions depends upon the eddy shedding frequency, conventionally described by the Strouhal Number. The most commonly cited relationship between Strouhal Number and Reynolds Number for circular cylinders was developed by Lienhard [1], whereby the Strouhal Number exhibits a consistent narrow band of about 0.2 (conventional across the sub-critical Re range), with a pronounced hump peaking at about 0.5 within the critical flow regime. The source data underlying this relationship is re-examined, wherein it was found to be predominantly associated with eddy shedding frequency about fixed or stationary cylinders. The pronounced hump appears to be an artefact of the measurement techniques employed by various investigators to detect eddy-shedding frequency in the wake of the cylinder. A variety of contemporary test data for elastically mounted cylinders, with freedom to oscillate under one degree of freedom (i.e. cross flow) and two degrees of freedom (i.e. cross flow and in-line) were evaluated and compared against the conventional Strouhal Number relationship. It is well established for VIV that the eddy shedding frequency will synchronise with the near resonant motions of a dynamically oscillating cylinder, such that the resultant bandwidth of lock-in exhibits a wider range of effective Strouhal Numbers than that reflected in the narrow-banded relationship about a mean of 0.2. However, whilst cylinders oscillating under one degree of freedom exhibit a mean Strouhal Number of 0.2 consistent with fixed/stationary cylinders, cylinders with two degrees of freedom exhibit a much lower mean Strouhal Number of around 0.14–0.15. Data supports the relationship that Strouhal Number does slightly diminish with increasing Reynolds Number. For oscillating cylinders, the bandwidth about the mean Strouhal Number value appears to remain largely consistent. For many practical structures in the marine environment subject to VIV excitation, such as long span, slender risers, mooring lines, pipeline spans, towed array sonar strings, and alike, the long flexible cylinders will respond in two degrees of freedom, where the identified difference in Strouhal Number is a significant aspect to be accounted for in the modelling of its dynamic behaviour.


Author(s):  
Murilo M. Cicolin ◽  
Gustavo R. S. Assi

Experiments have been carried out on models of rigid circular cylinders fitted with three different types of permeable meshes to investigate their effectiveness in the suppression of vortex-induced vibrations (VIV). Measurements of amplitude of vibration and drag force are presented for models with low mass and damping which are free to respond in the cross-flow direction. Results for two meshes made of ropes and cylindrical tubes are compared with the VIV response of a bare cylinder and that of a known suppressor called the “ventilated trousers” (VT). All three meshes achieved an average 50% reduction of the peak response when compared with that of the bare cylinder. The sparse mesh configuration presented a similar behaviour to the VT, while the dense mesh produced considerable VIV response for an indefinitely long range of reduced velocity. All the three meshes have increased drag when compared with that of the bare cylinder. Reynolds number ranged from 5,000 to 25,000 and reduced velocity was varied between 2 and 15.


2013 ◽  
Author(s):  
Kintak Raymond Yu ◽  
Alexander Hay ◽  
Dominique Pelletier ◽  
Simon Corbeil-Létourneau ◽  
Shahin Ghasemi ◽  
...  

Vortex-induced vibration is an important phenomenon for offshore engineering. For applications like the piping in the deep water oil exploration projects, the mass ratios can be of order of one [1]. Hence, there is a practical need to understand the effects of low mass ratio on vortex-induced vibrations to enhance design safety. The main purpose of this study is to numerically explore the two degrees of freedom (transverse and streamwise) responses of vortex-induced vibrations of a cylinder at low Reynolds number for the limiting case of zero mass ratio and zero damping. We aim to characterize the responses. In particular, we focus on determining the maximum amplitude values. It is a continuation from the work of Etienne and Pelletier who studied such behaviors at very low Reynolds number (Re < 50) [2]. We investigate the responses in the following parameter space: Reynolds number (75 ≤ Re ≤ 175), reduced velocity (5.0 ≤ Ur ≤ 11.0) and mass ratio (m* = {0, 0.1, 1}) with a fully coupled fluid-structure interaction numerical model based on the finite element method. Our results are generally in accordance with those from previous works for the displacement trajectories, force phase diagram, and the trends in frequency response and oscillation amplitude. The maximum transverse amplitude is found to be around 0.9 in the studied parameter space. In particular, with zero mass ratio, the maximum transverse amplitude starts to occur at values of reduced velocity higher than 6.5 for Reynolds number larger than 150. This is in contrast to the results of Etienne and Pelletier [2] who found that the maximum transverse amplitude always occurs at the reduced velocity of 6.5 for Reynolds number less than 50. Furthermore, with zero mass ratio, the maximum transverse amplitude increases when the Reynolds number increases. This behavior differs from what was suggested by Williamson and Govardhan [3] for a cylinder oscillating only in the transverse direction at Reynolds numbers in the range of 85 to 200. They found that the Reynolds number has no influence on the maximum transverse amplitude. We do not notice any response branching in this parameter space. However, the results in the present work clearly consist of two distinct characteristics. This indicates that the investigated mass ratio values encompass the critical mass ratio; whose value is estimated to be around 0.1 to 0.2.


Author(s):  
J. R. Chaplin ◽  
W. M. J. Batten

The flow-induced vibration of one cylinder in the wake of another is the subject of continuing interest in connection with interactions between vertical tension risers in deep water. When one riser is downstream of another, it is likely to be subject to wake-induced and vortex-induced excitations at different frequencies simultaneously. Both are complex mechanisms, and it is reasonable to assume that they interact. To begin to understand this complicated process, it is desirable that any modeling should incorporate some features of a multidegree-of-freedom structural response. With this aim, this paper describes experiments in which one cylinder was free to undergo simultaneous wake- and vortex-induced vibrations downstream of a similar but stationary cylinder in a steady flow. The downstream cylinder was mounted on an elastic system that had two natural frequencies in both the in-line and cross-flow directions. Mass ratios were almost the same in all four modes. Measurements are presented of simultaneous wake- and vortex-induced vibrations for cylinder separations of 5 and 10 diameters in the in-line direction, and up to 4 diameters transversely. At a reduced velocity of 83 (based on the cylinder's lower submerged natural frequency) and a separation of 5 diameters, excursions of wake-induced vibrations peaked at almost 5 diameters, when the downstream cylinder was near the edge of the upstream cylinder's wake.


Author(s):  
Murilo M. Cicolin ◽  
Cesar M. Freire ◽  
Gustavo R. S. Assi

Experiments have been carried out on models of rigid circular cylinders fitted with three different types of permeable meshes to investigate their effectiveness in the suppression of vortex-induced vibrations (VIV). Measurements of the dynamic response are presented for models with low mass and damping which are free to respond in the cross-flow direction. Reynolds number ranged from 1,000 to 10,000 and reduced velocity was varied between 2 and 13. Also presented are measurements of the wake of static models with Particle Image Velocimetry (PIV) at Reynolds number equal to 4000. Results for two meshes made of ropes and cylindrical tubes are compared with the VIV response of a bare cylinder and that of a known suppressor called the “ventilated trousers” (VT). All three meshes achieved an average 50% reduction of the response when compared with that of the bare cylinder. The sparse mesh configuration presented a similar behaviour to the VT, while the dense mesh produced considerable VIV response for an indefinitely long range of reduced velocity. Visualisation of the flow by PIV around static cylinders revealed that all suppressors disrupt the vortex shedding and increase the formation length when compared to the bare cylinder. The VT mesh, which presented the best suppression, also presented the largest vortex formation length.


Sign in / Sign up

Export Citation Format

Share Document