A Simplified Method for Quantitative Reliability and Integrity Analysis of Steel Catenary Risers

Quantitative reliability and integrity analysis of steel catenary risers (SCRs) can provide important information about their safety and toward their cost-effective and optimal design. SCRs are one of the commonly used riser systems in offshore production stations. The consequence of an SCR failure is significant; however, the overall safety of the riser is typically not quantified. Especially, because of the uncertainties associated with environmental conditions and structural capacities, quantitative reliability methods can take advantage of available data and developments in computing technology to provide a strong basis for their reliable engineering decision making. This paper presents a simplified approach for assessing the strength and fatigue reliability of SCRs, accounting for the uncertainties with their yield strength and fatigue capacities as well as the environmental conditions. Moreover, the integrity-based optimal design of riser strength limit state for a target annual probability of failure is discussed. The fatigue reliability of the SCR system is also assessed in component and system levels. The proposed method is then applied to a typical SCR attached to a semisubmersible vessel under Gulf of Mexico (GOM) conditions. Results of dynamic (time-domain) analyses under various environmental conditions are used to quantify the SCR safety and integrity and to optimize its design for a target annual probability of strength failure. By estimating the riser system probability of strength and fatigue failure in its lifetime, the strength and fatigue integrity indices, and the optimality factors of the riser sections for the strength limit state, suggestions are provided to improve the riser design. For example, it was found that considering the two main limit states of strength and fatigue failure of the SCR system, a strength failure at the taper stress joint (TSJ) is the likely mode of failure in this riser system, which has a probability of 0.0035 in its 25 year lifetime.