Three-dimensional modelling of drag anchor penetration using the material point method

TL;DR

This work delivers a 3D Material Point Method–based tool for predicting drag anchor penetration in sandy seabeds, calibrated to CPT data. It introduces three key advances: assemblies of rigid bodies to model articulated anchors, a partitioned-domain strategy to handle long pull lengths efficiently, and improved modelling of rotational inertia, all validated against centrifuge tests. The CPT-calibrated framework accurately predicts anchor trajectories and ultimate penetration depths across a range of relative densities and anchor designs, and it exposes limitations in the CBRA framework, notably density-dependent seabed factors. The results enable site-specific anchor-penetration assessments along cable routes and offer practical guidance for anchor design and burial depth decisions.

Abstract

Drag embedment anchors are a key threat to buried subsea linear infrastructure, such as power/data cables and pipelines. For cables, selecting a burial depth is a compromise between protecting the cable from anchor strike and the increased cost of deeper installation. This presents an efficient large deformation, elasto-plastic Material Point Method-based soil-structure interaction predictive tool for the estimation of anchor penetration based on Cone Penetration Test (CPT) site investigation data. The tool builds on earlier work by the authors supplemented by three developments: modelling assemblies of rigid bodies (necessary for articulated anchors), a partitioned domain approach to enable accurate and efficient modelling of long anchor pulls and an improved means of modelling rotational inertia. The tool is validated against scaled physical tests conducted in a geotechnical centrifuge on sands with a range of relative densities with good agreement across the tested conditions. Numerical simulations identify key issues with the UK Cable Burial Risk Assessment (CBRA) approach for estimating anchor penetration and reveal the potentially non-conservatism of the CBRA framework for sandy seabeds. The numerical model enables site-specific anchor-penetration assessment along cable routes and can be used to evaluate the performance of different anchor designs and sizes in varied soil conditions.

Figures (27)

• Figure 1: Rigid body: (a) is a single line element with its normal $\bm{a}$ and tangent $\bm{r}$ being used to define the point $\bm{x}_{\Upsilon,i}^\prime$. (b) shows how two line elements are used to define the coordinates of the fluke and shank of an anchor.
• Figure 2: Rotational inertia: An example of a truss element with angular acceleration defined at the centre of mass $\bm{x}_A$.
• Figure 3: Limiting opening angle: When the opening angle is less than the limit the penalty spring is inactive (a) but when greater than a limit it is active (b).
• Figure 4: Partitioned domain: The mesh and initial material configuration for stage 1 is shown in (a) with the subsequent deformed material, new mesh and boundary conditions, and anchor (in red) for stage 2 shown in (b).
• Figure 5: Anchor solution and recording procedure: The three stages and their corresponding meshes with: (a) Settling of the material under self-weight; (b) Placement and settling of the anchor on the surface; and (c) Anchor pull and recording of fluke tip position.