Elsevier

Marine Structures

Volume 42, July 2015, Pages 154-174
Marine Structures

Bending behavior modeling of unbonded flexible pipes considering tangential compliance of interlayer contact interfaces and shear deformations

https://doi.org/10.1016/j.marstruc.2015.03.007Get rights and content

Highlights

  • The assumption is verified that wire slippage in its axial direction is only governed by the axial force and friction.

  • More rigorous analytical formulations are given to describe the gross slip initiation and progression of the armor wire.

  • Two mechanisms are thought to contribute to the decrease of layer stiffness before gross slip begins.

  • The first one introduces micro-slip of contact interfaces and the other considers shear deformations of the plastic layer.

  • Poisson's effect is found to have the influences of delaying the gross slip initiation and reducing the layer stiffness.

Abstract

Rigorous analytical formulations are given to describe the gross slip initiation and progression in tensile armor layers of unbonded flexible pipes. Then two mechanisms are thought to contribute to the decrease of layer stiffness before gross slip begins. The first one considers the micro-slip occurred at the interlayer contact interfaces. The relative displacement between an armor wire and the underlying layer is determined according to the theory of contact mechanics. Shear deformations of the supporting plastic layer are taken into account in the other mechanism where plane sections no longer remain plane. The results of bending moment-curvature relationship from the presented models are compared with the available test data and good correlations are found. The shear model is seen to describe the slip transition better than other models do.

Introduction

While flexible pipe behavior under axisymmetric loads is well understood [34], [15], [20], [2], [26], [25], [29], determination of various bending characteristics is still a challenging problem. The difficulty is mainly due to the presence of helical layers and the freedom of helical components to slide against surrounding layers. So many authors put substantial effort into the post-slip behavior modeling of flexible pipes [5], [35], [24], [27], [11], [12], [19], [25], [29], [4], based on the assumption that the slip path of the armor wire follows either the strained helix (also known as the loxodromic path) or a geodesic curve.

Experimental results in terms of bending moment-curvature relationship generally show much smaller pre-slip bending stiffness and longer transition from the pre-slip to the full-slip stiffness than the analytical prediction [33], which implies that the pre-slip behavior of flexible pipes is not described adequately. In current stick-slip models, the pipe is simply modeled as a solid beam before gross slip initiates, i.e. plane sections remain plane during bending. Recently shear deformation in the multi-layer structure is identified as a potential source of the stiffness decrease in the stick regime [28]. However, it is believed that tangential compliance of interlayer contact interfaces is probably another mechanism which plays some part in the reduction of the pre-slip stiffness.

According to the theory of contact mechanics [10], micro-slip does occur between contacting bodies up to a certain limit value at which gross slip begins. The relative tangential displacement is accommodated in large part by elastic shear strain at the contact interface. This elastic behavior is characterized by tangential compliance. Analytical models considering the tangential compliance of contact interfaces can be found in the literature, mostly dealing with the analyses of spiral strands or cables [9], [8], [7], [21], [18]. The inclusion of tangential compliance is translated into a slowly decreasing cable flexural rigidity [7], [21], [18]. To our knowledge, the consideration of tangential compliance in flexible pipe analyses has not been performed.

The pre-slip behavior has a significant effect on the fatigue performance of dynamic applications of flexible pipes, such as a deep-water flexible riser. In global analyses the decrease of pre-slip stiffness can give larger curvatures compared with the current stick-slip models. This effect is even more pronounced if bending hysteresis is incorporated [30], [31]. In addition, shear deformations will delay the gross slip initiation, hence reducing the stress in the stick domain at a given curvature [28]. Considering the slope parameter in the S–N curve, a small reduction factor in stress can have a great impact on the resulting fatigue life.

In this paper, more rigorous analytical formulations are firstly given to describe the gross slip initiation and progression of the armor wire. The effects of tangential compliance and shear deformations on the layer bending behavior are then taken into account. Comparative studies with respect to the available experimental data are further carried out and encouraging correlations are found.

Section snippets

Assumptions and definitions

The main assumptions and definitions used throughout the analysis are:

  • 1.

    The slip path of the armor wire is assumed to follow either the strained helix or a geodesic curve. Both curves are considered as the limit curves for wire slip, and the actual path is probably somewhere in between. According to [23]; the transverse slip can be neglected when evaluating the dynamic bending stress. Hence, the strained helix assumption is employed in this study.

  • 2.

    The wire slippage in its axial direction is only

Case study

Witz [33] published a case study in the cross-section analysis of a 2.5-inch unbonded flexible pipe. Both construction details of the pipe and the experimental results were provided for axisymmetric and bending behavior. The bending response of the pipe was measured under two different internal pressures, 0 and 30 MPa. Berge et al. [1] also reported the full scale bending test results of a 4-inch flexible pipe under the internal pressures of 0.7, 10 and 25 MPa.

Fig. 6 shows the boundary angles

Discussion on the TC model

The TC model stands on the relationship between the relative displacement and the tangential force, i.e. Equation (46) and Equations (49), (50). In the derivation of Equation (46), the contact dimensions and the pressure distribution due to the normal force are determined according to the Hertz theory and these quantities are assumed to be unaffected by the subsequent application of the tangential force [10]. Each individual asperity is also assumed to follow the Hertz theory in the model

Concluding remarks

By evaluating the shear forces on the wire cross-section, this study firstly verifies the commonly made assumption that the wire slippage in its axial direction is only governed by the internal axial force and the available friction. Then more rigorous analytical formulations are given to describe the gross slip initiation and progression of the wire. Two mechanisms are further thought to contribute to the decrease of layer stiffness before gross slip begins. The first one introduces the

Acknowledgments

Financial support from the National Natural Science Foundation of China (Grant No. 51409044), China Postdoctoral Science Foundation (Grant No. 2014M551088) and the Fundamental Research Funds for the Central Universities (Grant No. DUT14RC(3)005) is gratefully acknowledged.

References (35)

  • S. Berge et al.

    Handbook on design and operation of flexible pipes

    (1992)
  • H. Deresiewicz

    Oblique contact of nonspherical elastic bodies

    J Appl Mech

    (1957)
  • L. Dong et al.

    An analytical model to predict the bending behavior of unbonded flexible pipes

    J Ship Res

    (2013)
  • J.J. Feret et al.

    Calculation of stresses and slip in structural layers of unbonded flexible pipes

    J Offshore Mech Arct Eng

    (1987)
  • C. Hardy

    Analysis of self-damping characteristics of stranded cables in transverse vibrations

  • C. Hardy et al.

    On the dynamic flexural rigidity of taut stranded cables

  • C. Hardy et al.
  • Cited by (20)

    • Methodology for simulation of the post-failure behavior of flexible catenary risers

      2022, Marine Structures
      Citation Excerpt :

      Other factors, such as temperature, also retain a minor importance [56]. However, it should be noted that the definition of the moment–curvature relation is not trivial [51,57], and different researchers and companies employ distinct analytical or numerical methods to define the nonlinear behavior [51,54,57–59]. It is important to note that the assessment of the moment–curvature relation is not a focus of this paper, but rather the qualitative results it may provide.

    • Modelling of the hysteretic bending behavior for helical strands under multi-axial loads

      2021, Applied Mathematical Modelling
      Citation Excerpt :

      To author’s knowledge, the distribution of radial force in strands caused by bending load is a relatively unexplored field. Indeed, rigorous analytical methods can relative accurately formulate this problem [51–53]. However, due to complicated geometries and kinematics of overall strands, such methods could be overly complex and may lead to unexpected problems, e.g. Signorini problems [54].

    View all citing articles on Scopus
    View full text