For deepwater development in the Gulf of Mexico, steel catenary risers (SCRs) supported from both SPAR and semi-submersible platforms have proven to be successful solutions for in-field flowlines, tie-backs, and export systems. It is envisaged that this will continue to be a promising solution in ultra deep-water applications, up to and beyond 10,000 ft. The study, commissioned by the Mineral Management Service (MMS), investigated the reliability of large-diameter SCRs in ultra-deepwater operations. The primary damage mode considered is fatigue failure. A probabilistic methodology for fatigue reliability is developed, which utilizes deterministic cumulative fatigue damage indicators, namely the stress levels and cycles associated with the various sea states and the fatigue strength of the members. Uncertainties in structural load and material properties are accounted for by assigning probability distributions and standard deviations to the deterministic stress levels. Furthermore, fatigue strength parameters, Miner’s indices, and capacities are modeled as random variables. First order reliability method (FORM) is employed for estimating fatigue reliability. The methodology is applied to three deterministic case studies presented by Intec Engineering (2006a, 2006b). The case studies involved either a SPAR or a semi-submersible platform. For the sake of brevity, a case study involving only a SPAR platform is presented in this paper. The effect of uncertainties in parameters on fatigue reliabilities is investigated. It is observed that the fatigue reliability estimates followed similar trends as the deterministic cumulative damage results, and hence can be used to complement deterministic estimates. Additional benefit and insight gained from the probabilistic study, which can be used for design decisions, include information regarding probabilistic importance and probabilistic sensitivity analysis. For case study presented here, it is seen that in general, uncertainty in the fatigue strength exponent (m) has the highest impact on fatigue reliability of SCRs. The second most important random variable is the stress range (S), which captures uncertainties in parameters such as loads and material properties. Parametric sensitivity studies on the fatigue strength parameters indicate that SCR reliability is sensitive to both the standard deviation and probability distribution of the parameters, thus highlighting the need for accurate probabilistic calibration of the random variables.
Developing an Analytical Model and Computing Tool for Optimizing Lapping Operations of Flat Objects Made of Alloyed Steels