Experimental Study of Plate Anchor-Soil Interaction in Sand for the Development of a Novel Anchor

Abstract

Sustainable energy produced by offshore wind will likely increase as technology moves into deeper water. With increasing water depth, floating substructures may become the most economical and viable means for deploying offshore wind turbines, and thus require a greater reliance on anchoring systems. A green anchor concept called the “flying wing anchor” is currently in development to provide high vertical load capacity, and minimize the amount of energy needed to transport, install, and recover it from the seabed. Conceptually, the anchor is dynamically installed vertically through free-fall penetration, where the anchor will then rotate and dive into a position that is near normal to the anchor line in response to the service loads imposed by the offshore floating structure. To aid in the development of this novel anchor concept, an experimental program was conducted on scale-model anchors under 1g acceleration in a rigid sand-filled tank. Chapter 1 describes a novel laboratory approach to measure the strength (i.e. friction angle) at very low confining pressures, typical of 1g physical model experiments. A simple tilt method is proposed to capture the peak friction angle at very low confining pressures, and then combined with conventional triaxial results to calibrate a modified stress-dilatancy relationship. The results of the tilt method indicate adequate and rational estimates of friction angle at very low stresses, and show to predict the critical state friction angle of the test sands within 1% to 3%. Furthermore, the modified stress-dilatancy relationship minimizes the asymptotic nature of the standard relationship at very low confining pressures by adding a second logarithmic term. Chapter 2 presents an experimental and analytical study carried out in the same sand to investigate the effect of anchor shape on the pullout capacity of horizontal plate anchors. The experimental results indicate a difference in normalized capacity over a range of embedment depth with respect to shape. Circular anchors produced consistently larger capacities relative to square anchors, and with increasing embedment depth, circular, equilateral triangular, and kite anchors become comparable within 5%. The proposed analytical model predicted the pullout capacity within 10% for circular plates, and within, on average, 30% for the remaining shapes at shallow embedment depths. Chapter 3 presents a physical model study to identify the anchor shapes that have the most effective dive performance, and to investigate the effects of initial embedment depth, loading line location, initial fluke orientation, and loading line angle on the dive trajectory.