Carbon fibre buckle arrestors for offshore pipelines

Abstract

Feasibility and efficiency of using carbon fibre reinforced polymer (CFRP) buckle arrestors in steel offshore pipelines with D/t of 28 and 40 are investigated using hyperbaric chamber tests. CFRP arrestors are manufactured using Prepreg (PP), Wet-Layup (WL) and Vacuum Bagging (VB) curing methods, with coarse and fine sand surface preparations. A parametric study is performed that outlines the performance of CFRP arrestors in various geometric configurations. Efficiency of CFRP arrestors using different manufacturing methods and various geometric configurations are calculated and compared with those of conventional steel buckle arrestors. It is shown that the efficiencies of CFRP arrestors vary between 0.74 and 1.0 for different manufacturing methods. Optimum efficiencies are obtained in the WL technique, using fine sanding, with CFRP arrestor of thickness twice the steel pipe-wall thickness, and fibres oriented in the hoop direction. Results show that at similar efficiencies, the CFRP arrestors can be much thinner than conventional slip-on or integral arrestors.

Introduction

A major concern in design of subsea pipelines in deep and ultra-deep waters is the collapse under external hydrostatic pressure. In the presence of local damage in the pipe-wall (in the form of out-of-roundness, dents or corrosion) local collapse can be initiated in the pipeline. The hazardous subsea pipeline accidents caused by corrosion, 65% were because of external corrosion, many researchers have investigated the collapse of subsea pipelines due to corrosion (Zhang et al., 2016; Gamboa et al., 2008; Zhang et al., 2015; Zhang et al., 2016). Once local collapse initiated, the circular cross-section of the pipe transforms into a dog-bone shape, and eventually a flat shape (causing shutdown of the pipeline), as the buckle rapidly propagates along the pipeline (Albermani et al., 2011; Karampour et al., 2017; Alrsai et al., 2018; He et al., 2014; Alrsai et al., 2018; Yan et al., 2016). The corresponding local collapse due to external pressure may be coupled with other loadings in the pipeline, such as bending and axial force, resulting in buckle interaction (Karampour et al., 2013; Karampour and Albermani, 2016; Karampour and Albermani, 2014; Karampour et al., 2015; Karampour, 2018).

The lowest pressure required to perpetuate the local buckle is termed propagation pressure (P_P), which is typically only 15% of the collapse pressure (P_CO). In case the external pressure exceeds the propagation buckle criterion (DNV, 2017), buckle arrestors are installed at certain intervals along the pipeline based on cost and spare pipe philosophy. Such arrestors are snuggly fitted around the pipeline to limit the damage and safeguard the downstream section of the pipeline (Netto and Estefen, 1996; Kyriakides et al., 1998; Lee and Kyriakides, 2004).

Existing buckle arrestors are made of stiff metal rings which locally augment the circumferential stiffness of the pipeline, and thus hinder the buckle propagation. A buckle arrestor can halt the buckle completely, or may allow the buckle to cross-over at a higher pressure. The buckle arrestor pressure capacity (P_X) is closely related to the length L, thickness h and yield stress σ_ya of the arrestor as well as diameter D, wall thickness t, and yield stress σ_y of the pipe . The efficiency of a buckle arrestor (η), is defined as (Kyriakides and Babcock, 1979)

$$\eta =\frac{P_X-P_P}{P_{CO}-P_P}$$

where P_P and P_CO are the propagation and collapse pressures of the adjacent pipe, respectively. An efficiency of 1.0 (P_X = P_CO) is achievable, if the buckle can be arrested in the upstream section of the pipeline. Most common types of buckle arrestors are: (1) slip-on arrestors (Lee and Kyriakides, 2004), where the arrestor is slipped over the pipe, (2) integral arrestors (Kyriakides et al., 1998), where the arrestor is welded to the pipe, (3) spiral arrestors (Kyriakides and Babcock, 1982), in which the arrestor is wound onto the pipe, and (4) clamped arrestor (Kyriakides, 2002). Amongst those, slip-on arrestors and integral arrestors are more prevalent. Slip-on arrestors are normally preferred to integral arrestors, since no welding is required. However, previous research has shown that their efficiency is normally lower than the integral arrestors (Kyriakides, 2002). The integral arrestor is a thick ring that is welded onto the pipeline. The weld should be robust enough to resist large deformations during the buckle propagation (Kyriakides et al., 1998; Yu et al., 2017). Therefore, the costs of installation of integral arrestors are significantly higher than other options.

Due to its excellent properties, such as high specific strength and stiffness, performance to weight ratio, thermal stability and corrosion resistance (Wonderly et al., 2005; Keller et al., 2013; Goertzen and Kessler, 2007; Sen and Mullins, 2007), carbon fibre reinforced polymer (CFRP) wraps are used to repair corroded and mechanically damaged offshore pipelines (Shamsuddoha et al., 2013; Duell et al., 2008; Seica and Packer, 2007). Moreover, experimental and numerical investigations have proved that the application of CFRP in pipeline repair improves the capacity of the damaged pipeline in carrying bending, compression, tension and torsional loads, in both quasi-static and cyclic loading scenarios (Lukács et al., 2010; Lukács et al., 2011).

The common buckle arrestors may hinder the pipe laying operation. For instance, in reel-lay method the pre-installed devices (such as buckle arrestors) interfere with the reeling and unreeling process (Lukács et al., 2010). Moreover, current buckle arrestors cannot be used in the inner components of pipe-in-pipe systems and pipeline bundles (Karampour et al., 2017; Alrsai et al., 2018; Olso and Kyriakides, 2003; Alrsai and Karampour, 2016). The current study proposes a CFRP buckle arrestor and investigates its feasibility, efficiency and appropriateness in offshore pipelines. The current work complements a previous study by the current authors (Karampour et al., 2019) which proved the feasibility of the CFRP buckle arrestor concept. To do so, experiments are conducted on stainless steel pipelines with diameter-to-thickness ratio, D/t, of 28 and 40, without arrestors (bare samples) and with CFRP arrestors, in a hyperbaric chamber. A parametric study is conducted to find the optimum thickness (h), length (L), and orientation (θ°) of the CFRP arrestor. Moreover, different CFRP arrestor manufacturing methods are tested. Using the experimental results, efficiency of CFRP arrestor is calculated and compared against those of existing buckle arrestors.