Fatigue analysis of offshore piping systems

Abstract

Oil and gas fields located in offshore are today being developed in even harsher and more challenging environments than anyone had thought of before. New designs, technologies, regulations and requirements have been developed and implemented along with these changes. As a result of these harsh conditions, the offshore structures will experience a lot of challenges in terms of design and maintenance integrity. One of the most important concerns is the wave loadings which are critical on offshore structures in these environments due to their cyclic behaviour over time. The structures considered in this thesis are pipe lines, which are influenced by wave loadings. The wave loadings considered in this thesis are high cyclic loadings, which will accumulate damages on structures and then lead to fatigue failures. These failures are a result of a combination of the stress amplitudes and the number of cycles. ASME B31.3 is the piping code that is utilized in design of most offshore process piping systems. But due to its lack of information about high cyclic fatigue failures, other codes need to be considered on this matter. There are different specifications which address fatigue failures, and the code used in this thesis is PD5500 British standard specification. This is used as a reference approach to estimate fatigue life. As an experiment there are two different other approaches discussed. One is covering the fatigue by calculating equivalent stress range and the other is covering the fatigue by assuming that a probability density function of stress range may be represented by a two parameter Weibull distribution. Examples from the Eldfisk and the Snorre fields have used for explanations of above given approaches. One has been the bridge piping on the Eldfisk field and the other one has been the flow line on the Snorre A field. The approach given in the PD5500 specification and the equivalent stress range approach gives the same results for the fatigue life estimation, but the third approach which assumed a two parameter Weibull distribution of stress range, gives a different result than the other two. The equivalent stress range approach can be proven analytically, but hasn’t proven earlier to be used with the two slopes SN curves. The thesis is discussed about the feasibility of using the equivalent stress range approach as another way of high cyclic fatigue assessment. This approach can be suitable to use in the industrial fatigue analyses but not the third approach. Fatigue damages on offshore topside piping systems in the North Sea have been evaluated by using the above examples and it has identified that more than 80% of the fatigue damages happened at the wave heights between 2 m to 8 m.