No abstract

Anchor flanges are interface items which are used to connect pipelines to subsea in-line and end termination structures. They are forged and tend to be long-lead items; therefore, the design of an anchor flange should be completed at a very early project stage, possibly even during the tender phase. An optimised, analytical method for preliminary design would result in reduced design time overall and have beneficial cost implications. The analytical notch methods (i.e. Neuber’s and Glinka’s) that are presented utilise linear-elastic stress concentration factors to make realistic predictions of the ultimate load carrying capacity of an anchor flange in the non-linear regime. The linear-elastic stress concentration factor values are calculated with simple analytical formulae and graphs. The analytical notch methods are deployed to predict the anchor flange limit load and peak plastic strain and thereby ensure that the plastic strain remains within the allowable limits of design codes. The cost and time saving associated with the analytical notch methods, and the accuracy that is maintained, are assessed by comparison with the predictions results obtained from detailed finite element analyses.

The Electrically Heat Traced Flowline (EHTF) is characterised by a combination of high performance dry annular thermal insulation of Pipe-in-Pipe (PiP) with a restricted electrical heating capability provided by helically wound copper wires laid between the inner pipe and the insulation in the annulus. The main advantage of EHTF are: future tie-back integration, unlock marginal reserves, access to HPHT pipeline, extend field life and maximise economic recovery and reduction in chemical and energy usage operational flexibility in controlling the flowline temperature and preventing the formation of wax and hydrates in shutdown conditions. Fibre optic cables are deployed in the EHTF system to measure the temperature of the flowline. This paper presents the development of a detailed finite element model to predict the mechanical behaviour of the helically wound cabling during reeling operations. The wires and cables were represented explicitly in the model as initially straight and then wound helically around the inner pipe with specified pre-tension. The EHTF PiP system was then cyclically deformed against a former to simulate the reeling process. A fibre optic cable (FOC) containing a local imperfection due to denting was included in the model to assess the impact of reeling process on the continued acceptability of accidentally dented FOC. The effects of friction between the cabling and the inner pipe and insulation surfaces, the pre-tensioned helical winding process and helix pitch, and the restraint provided by the thermal insulation layer and centralizers, were all investigated. Physical tests were conducted to establish the cyclic material properties of the electrical wires and results from these tests were used to calibrate the FE model. This paper details Subsea 7's technical expertise in modelling the highly complex behaviour of the EHTF cabling system as it experiences multiple bending cycles due to reeling. The paper highlights some important key results describing the behaviour of the wires and consequent predictions of integrity which have since been verified through full scale physical tests. The FE modelling also contributed to the insight gained regarding the overall behaviour of the system.

Pipelines are coated with insulating material that minimizes heat losses to the environment. Reeled pipe can experience nominal bending strain in the order of 1% to 2%. Thick coating on the pipe is inherently more highly strained, because of concentrations that occur at the interface between parent coating and field joint coating. Occasionally, contractors who specialize in pipe-lay using the reeling method have experienced difficulties relating to unexpected disbondment and cracks in coating at these interfaces. Any disbonded coating is routinely identified and repaired, but it is important to understand the influential factors that could lead to this type of coating disbondment. It is known in the industry that parameters such as temperature, reeling speed and pipe tension are influential but the relative influence of the factors is not well understood. In addition, there is currently no industry code or recommended practice that proposes the strain levels that the coating could safely withstand prior to cracking. This paper addresses thermo-mechanical aspects of coating design and presents a novel approach to quantify which parameters have the largest influence. In the presented assessments, coating strain was assessed using finite element analysis. Material input was selected from a combination of typical values and specific laboratory test results for polypropylene (PP) and injection molded polypropylene (IMPP). An essential aspect was that the mechanical and thermal properties of the PP were related to temperature and strain rate. Strain rates in the coating during reeling operations were obtained from global FE models. Detailed local FE models incorporated all the material and load inputs and temperature conditions that are necessary to determine peak strain values in the coating; the peak strain values would indicate the locations of potential coating disbondment. The study is purely a strain assessment and excludes any potential for defects or delamination in the coating that could result from its manufacturing process. This strain-based study revealed that coating temperature during reeling is the most influential factor on strain level in the coating. Reeling speed and pipe tension are parameters providing secondary influences.

In general, aeroegine casings may experience an axial force, a bending-moment and radial loading. Under these loads, the high stress regions of these complex aerongine casings will experience local stress and strain concentrations, with various load combinations. The stiffness will also depend on the loading mode. Hence, careful design is required to avoid the various types of failure such as buckling, crack initiation and propagation must be taken into account when designing an aeroengine casing structure. In addition, aerongine casings require extremely high reliability in service and adequate strength under extreme load conditions, i.e. Fan-Blade-Off (FBO) condition, must be demonstrated. Under radial loading of aeroengine casings, which have spoke to shell connections, these are the most likely sites for plastic deformation to occur and cracks to initiate. Also, the load path for each spoke to shell connection within the casing structure changes during loading. Based on these observations, this paper concentrates on the behavior which occurs in spoke to shell connections, referred to as local joints. The intention is first to characterize the local joint behavior and then to incorporate this into a global casing model. The work reported in this paper includes studies of mesh sensitivity, predictions of load path at each local joint under radial load, FE failure loci, upper bound techniques for predicting limit loads and stresses-strains predictions at local casing notches under elastic-plastic and creep situations using approximate notch methods. Hence, the global responses of a casing structure were predicted by utilizing a “repeated local joint” technique in conjunction with simplified global models.