The purpose of the work described here is to analyse vortex induced vibration VIV) effects on marine risers and unorthodox forms of suppression of this phe¬nomenon. which can cause structural failure through metal fatigue. Two suppression methods are explored: flex joints and buoyancy modules. Flex joints. which act like a hinge at a point on the riser due to the large reduction in bending stiffness. can suppress higher modes of vibration from propagating along or appearing in the riser. Buoyancy modules, with their local 100% increase in riser diameter. can decorrelate vortex shedding along the span of the riser and reduce the resonant effect of VIV.
The numerical finite element cable model and rigid body model developed at the University of Victoria were modified and used as a foundation for the research. The modifications include an algorithm to estimate the forces clue to ocean surface interaction with rigid bodies and a model to produce the appropriate VIV response in the numerical cable model. The resulting VIV model was calibrated and validated
with analytical. experimental, and numerical data available in the literature. In general. the model produces qualitative effects of VIV. including its self-starting and self-limiting nature, frequency lock-in. multi-mode response. and limited structural response on the order of one diameter.
A simulation of a testbed riser in a variety of ocean currents was generated to observe the effects of installing flex joints and buoyancy modules at various locations along the riser span. The performance of the testbed riser was gauged by comparing the time series of von Mises stress and the associated safety factor, ns. from fatigue failure at many points along the span to an unmodified testbed riser.
The stress fluctuation was drastically reduced within the flex joints for all water currents studied, which greatly increases fatigue performance. Flex joints placed at the top of the testbed riser had less impact. as the stresses are dominated by the large and unavoidable tensions found there. Flex joints placed in the bottom region of the riser did not affect the ns,. of the remaining riser span until very high modes of vibration were present. At these higher modes. some testbed riser configurations changed their vibration envelope and frequency. which indicates that a possible alternate and less damaging mode of vibration was induced. Flex joints therefore act effectively as a local patch against poor fatigue performance and placement of several flex joints does not negatively impact the behaviour of the rest of the riser in the cases examined. However. the explicit relationship between placement and spacing of flex joints with environment conditions remains unknown.
Buoyancy modules introduced spatial fluctuations in the entire nu profile of the testbed riser, unlike flex joints. In addition. the buoyancy modules decreased n, performance due to the hydrodynamic load concentrations induced by their large diameters. However, the 16% coverage case increased n.,. elsewhere along the riser, though the 10% covered riser did not match this performance. Since in both cases the modules were evenly spaced along the riser. performance benefits from increased coverage implies a minimum coverage of 16% needed for significant improvement in
fatigue performance for devices of this type. This coverage requirement may apply to traditional VIV suppression devices such as helical strokes. since they decorrelate vortex shedding along the span of the riser albeit in a different manner than buoyancy modules. Finally. the buoyancy modules changed the stress oscillation frequency more than the flex joint cases from the unmodified riser. This is desirable since lowering the frequency of oscillation also increases the fatigue performance of the riser.
Identifer | oai:union.ndltd.org:uvic.ca/oai:dspace.library.uvic.ca:1828/2342 |
Date | 10 March 2010 |
Creators | Nicoll, Ryan Stuart |
Contributors | Buckham, Bradley Jason, Suleman, Afzal |
Source Sets | University of Victoria |
Language | English, English |
Detected Language | English |
Type | Thesis |
Rights | Available to the World Wide Web |
Page generated in 0.0016 seconds