Plate anchors for offshore foundations in sand and clay

  • Joseph Fanning

Student thesis: Doctoral ThesisDoctor of Philosophy

Abstract

As the offshore wind industry moves into areas of deeper water, traditional foundation structures such as monopiles will become less economically feasible. This has generated a push for the development of floating turbines, which are held in place by mooring lines with anchor foundations in the seabed. As a result, the development of anchor foundation concepts, which can resist large vertical and horizontal loads have become a focus of research in recent years. Due to the varying ground conditions encountered in the offshore environment, various types of anchors have been developed. This thesis reports the findings of laboratory based investigations into the installation and loading behaviour of two novel plate anchor foundation concepts.

The concept for a Bi-wing Anchor which can be dropped through the water column and penetrate into the seabed under its own weight was introduced. After dropping, the anchor could then be dragged, causing it to embed further into the seabed. This would allow the anchor to be installed more quickly and to greater embedment depths than conventional drag-in plate anchors.

Physical modelling was used to analyse the drag behaviour of this anchor. Tests were completed in a transparent clay surrogate, which was consolidated through the application of a pressure gradient across the sample‟s depth. The observations from these tests were used in the development of an analytical model for drag embedment. The analytical predictions showed good agreement with the observed anchor behaviour at multiple embedment ratios.

The ultimate capacity of the anchor during pullout was analysed through centrifuge modelling in normally consolidated kaolin clay. The effect of the anchor‟s shape on its pullout capacity was analysed by varying the space between two anchor flukes in a series of tests. The anchor shape had minimal effect on deep and intermediate soil failure mechanisms. For shallow failure mechanisms, increasing the spacing between the flukes resulted in an increased normalised anchor capacity. Due to the uncertainty involved in predicting the final embedment orientation of drag embedment anchors, tests were then carried out to determine the effect of load angle and fluke angle on the pullout capacity. Capacity decreased as the inclination of the anchor fluke moved from a horizontal orientation towards a vertical orientation. The capacity also decreased as the direction of loading moved away from an angle normal to the fluke.

The implementation of dynamically installed anchors in granular soils has been proven to be problematic by previous studies. An alternative plate anchor concept, referred to as the Umbrella anchor was proposed for use in granular soils. This anchor could be pushed into soil deposits in a folded arrangement to reduce installation loads. Application of a pullout load to the anchor‟s mooring line would then cause the anchor to open, creating an embedded plate. Tests were carried out in saturated sand with the anchor installed at depths of up to 1.8m below the surface. The anchor could be installed quickly and accurately to a specified depth, in a short period of time. The opening of the anchor was observable in the loaddisplacement behaviour and confirmed by excavation of the anchor after testing. The normalised pullout capacity at multiple embedment ratios was comparable to the capacity obtained through wished-in-place tests and previous studies of circular plate anchors in sand.

The observed behaviour for both anchor concepts provided encouraging results and suggests that with further development and analysis, both could potentially be used for commercial applications.
Date of AwardJul 2020
Original languageEnglish
Awarding Institution
  • Queen's University Belfast
SponsorsUK Research and Innovation
SupervisorDr David McKee (Supervisor), Vinayagamoothy Sivakumar (Supervisor) & Shane Donohue (Supervisor)

Keywords

  • sand
  • clay
  • model test

Cite this

'