An investigation into the keying behaviour and the capacity of plate anchors in sand

Abstract

This thesis considers the potential of plate anchors as an anchoring option for offshore renewable energy devices such as wave energy converters and floating offshore wind turbines. In this study the performance of a plate anchor vertically installed in sand and subjected to vertical loading has been investigated experimentally. Particular focus was placed on the unrecoverable loss of embedment during the keying process, where the orientation of the plate evolves from vertical to perpendicular to the direction of loading. This is particularly significant for offshore plate anchors as an unrecoverable loss in anchor embedment corresponds with a loss in potential anchor capacity. The loss in embedment during keying was examined for six anchors, all with the same plate geometry, but with anchor padeyes (or load attachment points) that were at differing eccentricities from the plate. The experiments were conducted at model scale using the geotechnical centrifuge at the Institute of Technology Sligo. To facilitate observation of the anchor orientation and quantification of the loss in embedment during the test, anchor tests were conducted adjacent to a Perspex panel on the centrifuge strongbox. Vertical loading was achieved by pulling a mooring line attached to the anchor padeye at a constant velocity. The location and orientation of the anchor during each anchor test was captured using a high resolution digital camera mounted directly in front of the Perspex panel. The experimental data show that the loss in embedment of the plate anchor during keying is inversely proportional to the padeye eccentricity, with a padeye eccentricity equal to at least the breadth of the anchor plate giving minimal loss in embedment and hence highest potential anchor capacity. The magnitude of the loss in embedment is very similar to previous findings for clay. The peak anchor capacity was observed before the end of keying, at a plate orientation between 50 and 80 degrees to the horizontal. Particle image velocimetry was employed to reveal the failure mechanisms during the keying process. These analyses showed that the peak load corresponds with a sudden transition from a deep localised failure mechanism to a shallow mechanism that extends to the soil surface. The anchor capacity, expressed in terms of a dimensionless capacity factor, was shown to be in good agreement with previously reported experimental data on pipelines and strip anchors, but only after the peak anchor capacity is exceeded and the anchor behaves like a horizontally oriented anchor subjected to vertical loading. The particle image velocimetry analyses show that the inclination of the slip planes in the shallow failure mechanism are at an angle that is much lower than would be reasonable for a mobilised friction angle. This clearly shows that the normality condition, in which the dilation angle and the friction angle are equal, was not met in these tests and explains why the experimental data are in good agreement with predictions from a limit equilibrium method based on similar principles.