Search for a command to run...
Pipelines are widely used in the oil and gas industry in both offshore and onshore operations. After several years of operation in more or less corrosive environments, existing steel pipelines may suffer from internal or external metal loss due to erosion and/or corrosion damage mechanisms. More than 60 percent of the world’s oil and gas transmission pipelines are more than 40 years old and for the most part in urgent need of rehabilitation in order to re-establish the original operating capacity. The most recent repair method used for the repair of internally or externally corroded pipelines is a method based on utilising fibre-reinforced polymer (FRP) composites. This involves wrapping the corroded part of a pipeline with this material. Many laboratory tests and field experiments have been performed and documented in the past few years, which have proven the viability of this method. Two design codes are available for the design of FRP overwrap repairs in the rehabilitation of pipeline systems. These codes are based on the deterministic approach for design, which has been the common method of developing codes in the past. On the other hand, the more modern methods of design, such as the ultimate state design method include safety factors, which are probabilistically determined using reliability-based analyses for a target reliability factor. This dissertation critically analyses and reviews the two major pipeline repair design codes by means of analytical and finite element analysis. This is then followed by a proposal for a reliability-based design method for FRP overwrap repair. Chapters 5 to 7 present the work carried out on the reliability-based design method. The following steps have been followed in the reliability study: • Preparing large number of GFRP specimens by having them post-cured in two different environmental conditions; one set post-cured at 60◦c dry condition and the second set immersed in 6o◦c seawater for about 4 months. The first condition replicated the environment of the onshore pipelines and the second, the sub-sea pipelines; • Performing comprehensive tensile testing of the GFRP specimens in order to find the mechanical properties of GFRP laminates; • Fitting statistical models to the experimental data obtained having the best-fit models chosen as the representative models; • Determining the representative statistical model for different loads and resistance parameters; • Formulating the limit state function and setting the safety factors; and • Selecting the reliability index, running the reliability analysis and calibrating the safety factors. The end product of this research is a probabilistic design method. This proposed method allows the design of FRP overwrap repair based on the Tsai-Hill equation and uses two safety factors, one for the steel pipe and the other for the FRP repair. The steel partial safety factor is constant and equals 0.72, but the FRP partial safety factor varies with the depth of corrosion. Two graphs are presented to calculate the FRP safety factors, i) for onshore pipelines and ii) for offshore pipelines.