Elsevier

Marine Structures

Volume 43, October 2015, Pages 107-124
Marine Structures

Reliability-based design of subsea light weight pipeline against lateral stability

https://doi.org/10.1016/j.marstruc.2015.06.002Get rights and content

Highlights

  • A dynamic model is used to address the stability analysis of pipeline with considering fluids-pipe-soil interaction.

  • Using FEM calculation results, a response surface model (RSM) is developed to predict the pipeline displacement.

  • Combing with the RSM equation, the Monte Carlo simulation method is used to calculate the safety indices.

  • A calibrated factor is introduced into the serviceability limit state (SLS) function to calculate the failure probability.

  • Additional weight required for the on-bottom stability can be determined to meet the target levels of failure probability.

Abstract

The application of non-metallic light weight pipeline (LWP) in subsea oil/gas transmission system is subject to subsea pipeline on-bottom stability problem because of their light weight. Additional weight required for the stabilization of subsea LWP is a critical item to consider when decreasing the cost of the pipeline system. This paper presents an effective approach to determine the additional weight by utilizing a reliability-based assessment of subsea LWP against on-bottom stability. In the approach, a dynamic non-linear finite element model (FEM), including a model of fluids-pipe-soil interaction for the subsea pipeline, is used to study the pipeline displacement response. In-place analysis of a flexible pipe is presented as an example of the authors' methodology. Results show that displacements are largely affected with and without considering the lift force. Additionally, the uncertainties of all parameters used in the model are considered. With 145 cases of FEM calculations being the samples, a response surface model (RSM) is developed to predict the pipeline lateral displacement using the software Design-Expert. Combing with the RSM equation, the Monte Carlo simulation method is employed to estimate the probability of exceeding pipeline stability. To calculate the reliability of LWP for different submerged weights, the method introduces a calibrated factor into the serviceability limit state (SLS) function. The proposed approach can be used to determine the additional weight required for the on-bottom stability of subsea pipelines while considering the uncertainties of all relevant parameters.

Introduction

With increased corrosion failures and the degradation of rigid steel pipelines in subsea system for sour crude oil, anti-corrosion non-metallic light weight pipelines (LWPs) are considered to replace the rigid steel pipe due to their flexibility and ductility. However, one of the greatest challenges for the application of LWP to subsea systems is the on-bottom stability problem of the pipe because of its light weight [1]. The selection of additional weight for subsea pipeline on-bottom stability analysis is critical to decrease the cost of the pipeline system.

A comprehensive understanding of the fluids-pipe-soil interaction for subsea pipelines is very important to determine the additional weight required for the pipe on-bottom stability. Factors included are the hydrodynamic forces due to waves and currents (including lift, drag and inertia forces) and passive soil resistance [2]. The most common approach used in industry to assess the on-bottom stability of submerged pipelines is making conservative assumptions based on hydrodynamic loadings and soil resistance. However, the evaluation of hydrodynamic forces introduces large uncertainty into the analysis. For example, the universal force coefficients do not consider the effects of the body of pipeline-itself [3], [4], [5]. The evaluation of the interactive effects between the seabed and the pipeline is far more complex. The development of pipe-soil interaction models can be divided into three stages. Stage I: DNV RP-E305 [6] recommends conservative soil friction coefficients, usually, 0.7 for sand soil and 0.2 for clay soil to model the soil resistance. However, tests show that even for the same type of soil, friction coefficients show a considerable scatter [7], [8]. The recommended friction coefficients are selected based on the lower bound values and therefore may under represent actual values. Stage involves two-component empirical models that are proposed and developed based on the theoretical and experimental work [9], [10], [11], [12], [13]. The models consist of a basic coulomb friction component and a passive component. Yet for different models, the passive component depends on soil strength, unit weight of soil, embedment, loading history, etc. Stage III is the plasticity pipe-soil interaction model first introduced by Zhang et al. [14], [15], [16] and developed further by Tian and Cassidy [17], [18], Tian et al. [19] and Youssef et al. [23]. Compared to the previous models, the plasticity model provides a much more comprehensive understanding of the mechanics involved in the pipe-soil interaction. However, it is known that there are large uncertainties in governing soil parameters, load effects etc. in the model. Thus, it is still not a valid method regarding pipeline on-bottom stability.

The probabilistic analysis shall be used in the model when large uncertainties exist in the relevant parameters, where the uncertainties in the range of the basic variables are described statistically. The statistical values used to describe a random variable are the mean value and the coefficient of variation (CoV). These values may be obtained from recognized data sources in normal cases. It is then possible to use reliability analysis, taking account for all relevant variables and the uncertainties related to them. Once the probability functions of all parameters are obtained, the probability of pipeline failure due to on-bottom stability can be evaluated. The application of response surface model for the probabilistic analysis of pipeline on-bottom stability was first conducted by Youssef et al. [24].

Excessive lateral displacement of pipeline under wave and current loads may be considered to be a serviceability limit state (SLS) with a recommended target failure probability of 10−2 per pipeline per relevant period [25]. The target reliability levels have to be met in design to ensure certain level of safety. The aim of this paper is to provide an approach for the reliability-based determination of subsea pipeline additional weight allowance against on-bottom lateral instability. The paper first introduces a non-linear finite element model (FEM) that takes account of the fluids-pipe-soil interaction model. Then it performs a probabilistic analysis that takes into account all relevant parameters and the uncertainties in the FEM model. A response surface model in combination with the Monte Carlo simulation is used to calculate the safety indices and the required submerged weight can be determined to meet the target levels of failure probability.

Section snippets

Structural model

A non-linear FEM is constructed to model hydrodynamic loads on subsea pipelines resulting from irregular waves and currents. Fig. 1 shows a flow chart of the non-linear FEM for pipeline on-bottom stability analysis. The general non-linear FE software ABAQUS is used to model the fluid-pipe-soil system. The beam element PIPE31H is used to simulate the pipeline and the beam section of the pipe is determined by pipe diameter and wall thickness. The subsea pipeline is subjected to hydrodynamic

Case study

A flexible non-metallic light weight pipeline is selected for the pipe on-bottom stability analysis. Table 1 lists the input parameters used in the case study. The wave and current data are selected from a typical area in the South China Sea [27]. The soil type is calcareous sand. A finite element model representing a 500 m pipeline was configured using ABAQUS/Standard PIPE31H structural elements. A 500 m long pipeline is divided into 50PIPE31H elements. The pipe-soil interaction model is

Reliability assessment

Failure caused by excessive lateral displacement of the pipeline because of the loss of pipe on-bottom stability due to hydrodynamic loads may be considered to be a serviceability limit state (SLS). The limit state can be specified by the operator, guided by a tolerable displacement limit:g(Z)=R(Z)S

Where g(Z) is the failure function, Z is a vector of all uncertainty variables, R(Z) is pipeline design resistance, S is the serviceability limit state. According to structural reliability theory,

Pipeline submerged weight study

According to the case study in the last section, the pipeline lateral displacement will definitely exceed the design criterion. To determine what pipe submerged weight should have been required to meet the target probability level, pipeline cases with different submerged weight values using FE analysis were studied. Stabilization method such as trenching, anchoring, rock dumping and mattress have been used in the past to ensure the stability of the pipeline. It is assumed that in this paper the

Conclusion

Reliability-based analysis of subsea light weight pipeline against lateral stability based on a non-linear finite element model is introduced. In the FEM model, the hydrodynamic loads generated by irregular waves and soil resistance by pipe-soil interaction are included through Abaqus subroutines. An example of on-bottom stability analysis for flexible non-metallic light weight pipeline is selected as a case study. The analysis results show that pipeline displacement behavior varies widely

References (33)

  • A.H. Mousselli

    Offshore pipeline design, analysis, and methods

    (1981)
  • H. Brennodden et al.

    An energy-based pipe-soil interaction model

  • D.A. Wagner et al.

    Pipe-soil interaction model

    J Waterw Port, Coast Ocean Eng

    (1989)
  • R.L.P. Verley et al.

    A soil resistance model for pipelines placed on sandy soils

    J Offshore Mech Arct Eng

    (1994)
  • R.L.P. Verley et al.

    A soil resistance model for pipelines placed on clay soils

  • B.S. Youssef et al.

    Calibration of verley and sotberg soil resistance model for pipeline placed on calcareours soils

  • Cited by (7)

    • Dynamic Bayesian network-based reliability and safety assessment of the subsea Christmas tree

      2021, Process Safety and Environmental Protection
      Citation Excerpt :

      Using MCS, Marco N. Coccon (Coccon et al., 2017) conducted a risk assessment of an offshore structure, and Hyo-Nam Lee (Lee and Jang, 2015) predicted the reliability of a newly developed pyrotechnic pin puller. Yong Bai (Bai et al., 2015) presented an RSM to predict a pipeline lateral displacement. Several researchers (Cai et al., 2012a) proposed Markov processes and used them in offshore oil production systems, BOP, and safety instrumented systems (Chebila, 2020).

    • Sustainable crude oil transportation: design optimization for pipelines considering thermal and hydraulic energy consumption

      2019, Chemical Engineering Research and Design
      Citation Excerpt :

      Optimal pipeline design is a widely studied subject in the oil and gas industry, as the pipeline is an important facility to connect upstream, middle stream, and downstream entities of the oil and gas supply chain (Assis et al., 2019; Wang et al., 2019). Bai et al. (2015) studied the design of sub-sea lightweight pipelines and presented an effective approach to determine the additional weight. Tian et al. (2017) proposed a method for CO2 pipeline transportation design considering engineering practices and economic uncertainties.

    • Quantitative risk assessment of submarine pipeline instability

      2017, Journal of Loss Prevention in the Process Industries
      Citation Excerpt :

      Therefore, it is necessary to conduct a risk assessment for submarine pipeline instability, which is useful for risk management or decision-making of submarine pipeline instability. At present, a lot of efforts are made for the study of on-bottom stability of submarine pipeline, which includes experiments, numerical simulation and mathematical model etc. (Gao et al., 2006; Tang et al., 2016; Bai et al., 2015). But these studies mainly focused on the on-bottom stability design, and the recommended practice in standard file or finite element method are usually employed to analyze the pipeline stability.

    • A coupled FE model for flow-induced instability of the integrated pipe-soil system

      2018, 13th ISOPE Pacific/Asia Offshore Mechanics Symposium, PACOMS 2018
    View all citing articles on Scopus
    View full text