Seismic Design of Jumpers: The Coupling Conundrum

Subsea developments in seismically active areas present a number of engineering challenges. One such item, which is frequently underestimated, is design of jumpers to resist earthquake loading. Jumpers connect quite dissimilar pieces of subsea equipment, for example heavy manifolds on deep foundations, PLETs on mudmats and wellheads supported on slender well casings. If the structures connected have a different dynamic response (e.g. manifold, PLET and wellheads with different fundamental periods), the jumpers receive additional loads from the out of phase structure movements. Such “seismic coupling” is particularly an issue for high pressure high temperature (HPHT) fields, where stiff heavy jumpers are combined with large manifolds. This paper discusses possible approach to evaluate coupling, and provides practical advice for subsea designers. Specifically we address coupling from a point of view of the fundamental period of the connected structures and jumpers, making reference to results of dynamic modeling and codified experience from nuclear power plant design. Differentiation between cases which require fully coupled treatment, and those in which a more straightforward separation and analysis of the individual subsystems is very important to producing an efficient design in reasonable time frame. These aspects are illustrated using a case history from our project files. Coupled dynamic analysis of primary (supporting) and secondary (supported) systems may not be always feasible or desirable. The number of supported structures may be such that a coupled analysis may create computational difficulties. In addition to this, data regarding the secondary system may be not available at the time of the analysis of the primary system with the result that development of a coupled model would be not applicable. In cases where an attempt is made to develop complex coupled systems, sanity checks of such a system may be not an easy task. For these and other reasons, complete analysis of complex coupled systems is rarely performed. The objective of this paper is to explore the conditions under which an uncoupled analysis is justified and to give simplified indication to check the results of coupled models. The paper presents a case study of a manifold-jumper-PLET. The system is evaluated using a standard uncoupled approach in which seismic actions are computed for the manifold, jumper and PLET considered as separate structures. The resulting actions are compared to those from a “coupled analysis” in which the full system is analyzed. The standard uncoupled analysis is found to be unconservative in cases where the jumper’s natural period of vibration is similar to that of the manifold or PLET. This approach shows promise as a means of identifying cases where coupled analysis is required to correctly assess seismic forces.