Elsevier

Ocean Engineering

Volume 146, 1 December 2017, Pages 246-256
Ocean Engineering

Predicting jack-up spudcan installation in sand overlying stiff clay

https://doi.org/10.1016/j.oceaneng.2017.09.046Get rights and content

Highlights

  • Bearing capacity factor formula when spudcan penetrates sand over clay is proposed.

  • An analytical method could predict peak capacity for stronger underlying clays.

  • Guideline methods under-estimate spudcan resistance in sand over stiff clay.

Abstract

There has been a constant trend towards larger mobile jack-ups capable of operating for extended periods in deeper water and in harsher environmental conditions. This is increasing both the size of their spudcan footings and the operational bearing pressures. Though some new analytical methods to predict the load-penetration profile have been proposed, and shown to fit centrifuge experiments well, these methods were calibrated mainly using experimental and numerical data for sand overlying soft clay. This paper reports six centrifuge tests simulating spudcan installation in sand overlying a stiff clay. These are complemented with large-deformation finite element analyses, simulating the continuous penetration of the spudcan in sand over stiff clay. Modified Mohr-Coulomb and Tresca models were used to describe the sand and clay behaviour, accounting for the effects of strain softening on the response of the soil. These results provide confidence that a recently published analytical method to predict the peak capacity can indeed be extended to high bearing resistances and stronger underlying clays. Bearing capacity factors used to predict capacity in the underlying clay have, however, been updated to reflect the new database of results. A new formulae that explicitly accounts for increasing strength with depth profiles is provided.

Introduction

In water up to a depth of approximately 150 m, mobile independent leg jack-up rigs are widely used to perform most offshore drilling. The jack-up units are designed for more onerous conditions leading to increased spudcan bearing pressures. Although the popular rig classes exhibited maximum vertical installation bearing pressure in the range of 200–600 kPa, some of the reported field cases have indicated higher bearing pressures (see Fig. 1). Are the calculation methods used to predict the vertical installation of spudcans appropriate for these pressures and all offshore conditions?

For instance, the punch-through of a spudcan through a layer of sand into underlying clay has been intensively investigated in a geotechnical centrifuge by Craig and Chua, 1990, Teh, 2007, Lee, 2009, Teh et al., 2010, Lee et al., 2013a, Hossain et al., 2016 and Hu et al., 2014a, Hu et al., 2016. Table 1 summarises the testing database, with the shear strength of the clay at the sand-clay interface sum ranging from 7.2 to 25.8 kPa and the shear strength gradient in the range of 0–2.1 kPa/m. Because it is easier to lay in a centrifuge this database consists of soft clays. However, for many of the punch-through locations of greatest concern, such as in the South Baltic and the North Sea, layered sand over stiff clay stratigraphy with su > 40 kPa are common (Werno et al., 1987, Kellezi et al., 2005, Kellezi and Stadsgaard, 2012). There is a need to confirm the applicability of analytical methods in these stronger soil profiles. It is noted that the term “stiff clay” is used somewhat generically in this paper to describe clays of strength greater than those previously used in the centrifuge testing and numerical analysis on the sand-over-clay problem. These are clays greater than 40 kPa, though this is slightly different to that defined in ISO (2012).

Load spread and punching shear are the ‘standard’ methods included to estimate the peak spudcan bearing pressure in sand overlying clay in the ISO 19905-1 guidelines (ISO, 2012). There is mounting evidence that these methods produce low estimates of the spudcan bearing pressure during punch-through failure when the undrained shear strength of the clay is low (Teh et al., 2010, Lee et al., 2013a, Hu et al., 2015b). However, whether such underestimation of the spudcan bearing pressure also occurs for the higher underlying undrained shear strengths often encountered offshore requires further validation.

Recently, a new analytical penetration resistance-depth profile prediction method (hereinafter called the Hu et al. method) was proposed (Lee et al., 2013b, Hu et al., 2014b, Hu et al., 2015a). It was calibrated using a testing and numerical database with soft clay strength of 10–40 kPa at the sand-clay interface and strength gradient of 1.5–2.5 kPa/m (Lee et al., 2013a, Hu et al., 2014a, Hu et al., 2015a, Hu et al., 2016). In the prediction of the peak penetration resistance in the upper sand layer, the stress level and dilatant response, as well as the embedment depth, are taken into account using a discretised silo approach. In the prediction of spudcan capacity in the underlying clay, both the thickness of the trapped sand beneath the spudcan and the clay's undrained shear strength increment with depth are accounted for.

The motivation for this paper emanates from the need to characterise spudcan installation in sand over stiff clay and to validate analytical methods used to calculate load-penetration profiles. The term “stiff clay” is used for the clays with strength greater than 40 kPa in the following analyses. The main objectives of the testing and numerical programme are: (a) to model experimentally the penetration resistance of a spudcan of generalised geometry penetrating medium-dense sand overlying stiff clay in a beam centrifuge and to extend the existing testing database to incorporate clay of higher undrained shear strength; (b) to evaluate the performance of the guideline methods in estimating the spudcan penetration resistance in sand over stiff clay soils; and (c) to validate the Hu et al. method to predict spudcan penetration resistance-depth profiles. This will provide confidence for use of these methods in typical offshore conditions faced by practitioners.

Section snippets

Testing programme

Physical modelling of spudcan penetration in medium-dense sand overlying stiff clay was conducted using the beam centrifuge at the University of Western Australia (UWA). The centrifuge has a swinging platform with a standard rectangular strongbox. It has internal dimensions of 650 mm (length) × 390 mm (width) × 325 mm (depth), representing a prototype testing area of 130 m × 78 m × 65 m respectively when testing at 200g. Tests were performed using four model spudcans of the same shape, but with

Testing results and predictions

The testing and numerical results are presented using prototype scale. The spudcan penetration resistance profiles for 6 medium dense sand overlying stiff clay centrifuge tests are shown in Fig. 6, Fig. 7. They are grouped for two thicknesses of the sand layer. The displacements are measured (zeroed) from the position of the bottom shoulder of the spudcan embedded at the soil surface (RP in Fig. 2b). The qpeak values are measured from the testing profiles in Fig. 6, Fig. 7. In general, all the

Analysis details and soil models

Apart from the testing data reported in this investigation, three-dimensional LDFE analyses were undertaken to simulate the continuous penetration of spudcans from the seabed, supplementing the testing database to a larger range of underlying stiff clay shear strength. The LDFE analyses were performed using the Coupled Eulerian-Lagrangian (CEL) approach in the commercial FE package Abaqus/Explicit. This large-deformation analysis approach provides a full spudcan penetration resistance profile

Generalised bearing capacity factor

During spudcan penetration in the underlying clay, the sand underneath is pressed and it forms a composite foundation with the moving spudcan. The bearing capacity factor Nc was derived by dividing the net penetration resistance by the undrained shear strength in the centrifuge tests and LDFE simulations, for which the Nc and su0 refer to the values at RP in Fig. 2b. Based on a large dataset of 111 cases, Hu et al. (2015a) derived an equation that fits all the data as:Nc=10.5+11HsD

When the

Conclusions

Six centrifuge tests have been conducted using a beam centrifuge to investigate spudcan foundation behaviour in sand overlying stiff clay. In contrast to previous soft clays testing database, the terminology “stiff clay” is adopted in this paper for the clays with strength greater than 40 kPa. The tests covered different prototype sand thicknesses of 4 and 6 m and spudcan diameters in the range of 6–12 m, corresponding to Hs/D ratios of 0.33–1. The CEL approach was used to replicate the

Acknowledgements

This work forms part of the activities of the Centre for Offshore Foundation Systems (COFS), which is supported by the Lloyd's Register Foundation as a Centre of Excellence and currently forms one of the primary nodes of the Australian Research Council Centre of Excellence for Geotechnical Science and Engineering. Lloyd's Register Foundation invests in science, engineering, and technology for public benefit, worldwide. This project received additional support from the ARC Laureate Fellowship (

References (29)

  • G. Qiu et al.

    Controlled installation of spudcan foundations on loose sand overlying weak clay

    Mar. Struct.

    (2011)
  • M.J. Cassidy

    Experimental observation of the penetration of spudcan footings in silt

    Géotechnique

    (2012)
  • M.J. Cassidy et al.

    A comparison of the combined load behaviour of spudcan and caisson foundations on soft normally consolidated clay

    Géotechnique

    (2004)
  • S.F. Chung et al.

    Effect of penetration rate on penetrometer resistance

    J. Geotech. Geoenviron. Eng.

    (2006)
  • W.H. Craig et al.

    Deep penetration of spudcan foundations on sand and clay

    Géotechnique

    (1990)
  • I.M.S. Finnie et al.

    Punch-through and liquefaction induced failure of shallow foundations on calcareous sediments

  • M.S. Hodder et al.

    Analysis of soil strength degradation during episodes of cyclic loading, illustrated by the T-Bar penetration test

    Int. J. Geomech.

    (2010)
  • M.S. Hossain et al.

    Extraction response of skirted foundation and a spudcan on sand-over-clay deposits

    Géotechnique

    (2016)
  • G.T. Houlsby et al.

    Undrained bearing capacity factors for conical footings on clay

    Géotechnique

    (2003)
  • P. Hu et al.

    Predicting peak resistance of spudcan penetrating sand overlying clay

    J. Geotech. Geoenviron. Eng.

    (2014)
  • P. Hu et al.

    Predicting the resistance profile of spudcan on sand overlying clay

    Can. Geotech. J.

    (2014)
  • P. Hu et al.

    Assessing the punch-through hazard of a spudcan on sand overlying clay

    Géotechnique

    (2015)
  • P. Hu et al.

    A comparison of full profile prediction methods for a spudcan penetrating sand overlying clay

    Géotech. Lett.

    (2015)
  • P. Hu et al.

    Effect of footing shape on penetration in sand overlying clay

    Int. J. Phys. Model. Geotech.

    (2016)
  • Cited by (21)

    • Advances in the prediction of spudcan punch-through in double-layered soils

      2021, Marine Structures
      Citation Excerpt :

      This method, for the first time, takes into account the stress level and dilatant response of the sand through an iterative procedure. The bearing capacity of spudcans in sand-over-clay was further investigated using centrifuge tests by Hu et al. [20,21]; Hu and Cassidy [28] and large deformation finite element (LDFE) analyses by Hu et al. [26,36]. Based on the new series of data by centrifuge testing and numerical modelling, a new ‘mechanism-based’ design method has been proposed.

    • Field experimental investigation of punch-through for different operational conditions during the jack-up rig spudcan penetration in sand overlying clay

      2020, Journal of Petroleum Science and Engineering
      Citation Excerpt :

      However, for the large deformation problem involving strong nonlinearity, complex contact conditions of soil and the nonlinearity of soil material, the singularity problem of computational mesh singularity is often encountered (Wu, 2014; Zhang et al., 2012; Zhou et al., 2017; Zhao et al., 2018). Several scientific research institutions have been conducting centrifuge model experiments, including the University of Western Australia, National University of Singapore, Tianjin University, Dalian University of Technology and China University of Petroleum (Hu and Cassidy, 2017). Teh et al. (2008) conducted centrifuge experiments in the sand overlying clay to further reveal the failure mechanism.

    View all citing articles on Scopus
    View full text