Full text
| |
| Author(s): |
Gustavo Roque da Silva Assi
Total Authors: 1
|
| Document type: | Master's Dissertation |
| Press: | São Paulo. |
| Institution: | Universidade de São Paulo (USP). Escola Politécnica (EP/BC) |
| Defense date: | 2005-12-16 |
| Examining board members: |
Julio Romano Meneghini;
Jose Augusto Penteado Aranha;
Sergio Hamilton Sphaier
|
| Advisor: | Julio Romano Meneghini |
| Abstract | |
This MSc Thesis presents an experimental study on flow-induced vibrations and interference effects around rigid circular cylinders free to oscillate transversally to the flow. Firstly, it justifies the real needs of an experimental approach within the context of offshore engineering, which motivates this project. After that, presents a review over bluff-bodies flows followed by some consideration concerning flow-induced vibrations. Special attention is found over the oscillations caused by VIV Vortex-Induced Vibrations and galloping phenomena, which are attributed to excite either a single isolated cylinder or a pair of tandem interfering cylinders. The experimental methodology is shown, describing the applications of elastic bases with one degree of freedom. Experimental techniques and signal analysis procedures are discussed considering the executions of these experiments in two water channel facilities. Presented, discussed and compared results involve: dynamic responses in amplitude and dominant oscillation frequency; instantaneous phase angle between fluid forces and cylinder displacement and phase angle betweens the two cylinders oscillations; and wake dynamics measurements and visualizations employing PIV technique. Cylinders are arranged as follow: single isolated cylinder; tandem pair with upstream one free to oscillate; tandem pair with downstream one free to oscillate; tandem pair with both cylinders free to oscillate. All carefully analyzed arrangements present low-mass parameter and very low damping . Gaps between cylinder centers vary through . Reduced velocity range is comprised in. The isolated cylinder case presented a typical VIV response, with three identified branches (initial, upper and lower). The instantaneous phase angle intermittency phenomenon was observed in transition regions. The 2S and 2P vortex modes were verified by PIV technique. These data showed to be in accordance to other literature measurements and are employed as reference results for comparisons throughout this text. The dynamic response of a cylinder oscillating upstream a fixed one showed a typical VIV behavior. From this can be concluded that the downstream cylinder does not imply any interference phenomenon for analyzed gaps and velocities. On the other hand, major interference effects occur when a downstream cylinder is oscillating in the wake of another fixed one. This configuration, which is the focus of this study, does not show a typical VIV response, since amplitude curves present a crescent branch after the typical synchronization regime. It does not either present a typical galloping excitation response, since the fluid forces are not in phase with the cylinder velocity. Thus, a new suggested name, WIV Wake-Interference Vibrations, describes the phenomenon responsible for these excitations, which combines: lock-in of shedding frequency from VIV; and asymptotically crescent response from galloping-like excitations. When both cylinders are free to oscillate, the upstream one presents the typical VIV response, while the downstream one passes through three branches with discontinuities, excited by WIV. Finally, it can be concluded that is not convenient to call galloping excitation" these crescent branches for interference arrangements. It can be understood that the dynamic response is being excited by the interference effects from upstream wake and downstream cylinder vortices. The nomenclature Wake-Interference Vibrations is suggested to be more adequate to the nature of this phenomenon. In addition, PIV visualizations showed to be very important to certify that the formation length decreases while increasing Reynolds number, requiring interference experiments with constant Reynolds numbers. To conclude, some suggestions for future work in this research field are presented. (AU) | |