Abstract
Steel Catenary Risers (SCRs) are widely used in the development of deepwater oil and gas fields due to their advantages of low manufacturing cost, resistance of high temperature and high pressure, etc. The vertical platform heave motions can cause significant vertical oscillations and hence relative fluid velocities close to the sag-bend area of an SCR. Vortex induced vibrations (VIV) can occur subjected to such periodic variation of relative flow speed.
In order to understand the mechanism of heave induced VIV, Statoil has carried out two model tests in the Ocean Basin in Shanghai Jiaotong University in 2012. A straight flexible beam was tested in an oscillatory flow in one of the tests to study the fundamental behaviour of VIV. Heave induced VIV was investigated in the other test, a truncated SCR model was tested with calculated motions at the top end. Recent studies on these experiments have revealed unique features of VIV in an oscillatory flow. The vortex shedding process is more complicated than that of constant flow. One of the reasons for this is that the pipe will interact with the recent shed vortices when it reverses its motion direction, which may amplify the response. VIV response is not only influenced by the reduced velocity, but also Keulegan–Carpenter (KC) number. In addition, different eigen-frequencies of the pipe can be excited and the mode transition is observed due to the variation of the flow speed during one oscillation period.
There is still lack of prediction tools to evaluate heave induced VIV. It can be desirable to modify the present empirical VIV prediction tools by including additional effects from oscillatory flow. In the present paper, the vortex shedding process during one oscillation period is approximated by several time spans with constant flows. VIV response at each time span is considered to be controlled by a representative flow speed. The final response will be the sum of the responses at different time spans. A riser system analysis program is used to determine the relative flow speeds along an SCR due to heave motions. Then, an empirical VIV prediction program is used to calculate VIV response at different time spans over one top motion period. The prediction results from this integrated approach are compared with experimental results.
In order to understand the mechanism of heave induced VIV, Statoil has carried out two model tests in the Ocean Basin in Shanghai Jiaotong University in 2012. A straight flexible beam was tested in an oscillatory flow in one of the tests to study the fundamental behaviour of VIV. Heave induced VIV was investigated in the other test, a truncated SCR model was tested with calculated motions at the top end. Recent studies on these experiments have revealed unique features of VIV in an oscillatory flow. The vortex shedding process is more complicated than that of constant flow. One of the reasons for this is that the pipe will interact with the recent shed vortices when it reverses its motion direction, which may amplify the response. VIV response is not only influenced by the reduced velocity, but also Keulegan–Carpenter (KC) number. In addition, different eigen-frequencies of the pipe can be excited and the mode transition is observed due to the variation of the flow speed during one oscillation period.
There is still lack of prediction tools to evaluate heave induced VIV. It can be desirable to modify the present empirical VIV prediction tools by including additional effects from oscillatory flow. In the present paper, the vortex shedding process during one oscillation period is approximated by several time spans with constant flows. VIV response at each time span is considered to be controlled by a representative flow speed. The final response will be the sum of the responses at different time spans. A riser system analysis program is used to determine the relative flow speeds along an SCR due to heave motions. Then, an empirical VIV prediction program is used to calculate VIV response at different time spans over one top motion period. The prediction results from this integrated approach are compared with experimental results.
