Modelling Seabed Interaction in Frequency Domain Analysis of Mooring Cables

Paul Pang Awn Ong

Abstract:

With the increasing demand for energy, the supply of offshore oil and gas has also beome increasingly important. The exploration and production of such a reserve often require the use of a floating vessel, anchored to the seabed by mooring cables for safe operation. The analysis of these cables is therefore a challenging one, and this is due in part to the cable's interaction with the seabed.

This cable-seabed interaction can be resolved into two primary actions. First, the interaction involves the axial stretching and relaxation of grounded cable (i.e. the section of cable that lies on the seabed). Second, it involves the lifting off and touching down of the cable on the seabed, which we term here as catenary action. Traditionally, it is believed that the latter can only be modelled accurately by time integrating the cable's equations of motion. This is far too computationally expensive for the many load cases that have to be considered.

A more efficient alternative would be to analyse the cable in the frequency domain, where the cable is assumed to undergo simple harmonic motions. Such an assumption also means that seabed interaction would have to be simplified. Up until now, this has been achieved either by pinning the cable at its touchdown position, thus ignoring the interaction altogether, or by modelling the grounded cable with an equivalent horizontal spring. Neither of these methods offer a proper way of modelling the catenary action mentioned above.

A new method is therefore proposed. In this new method, seabed interaction is modelled by a system of coupled linear springs, rather than a single spring. The cable is truncated at an optimum position near the seabed and the truncated section replaced by a system of springs. Static catenery equations are then linearised, following a procedure routinely adopted in the linearisation of fluid drag, to yield the stiffness values for the spring system. Doing this, it is assumed that the spring system would replicate the behaviour of the truncated line, and hence model the interaction concerned.

The proposed method has shown to improve the accuracy of frequency domain analysis in a number of cases. Cables of different pretensions found in three broad categories of water depths - shallow, deep and ultra-deep - have been analysed in both frequency and time domains. Where available, model test solutions have also been used as benchmark. The proposed method has shown the greatest improvement for large excitations of low pretension cables, or those deployed in shallow waters. For depths of 1000 m or more, seabed interaction has no effect on the cable and can thus be modelled simply by pinning the cable at its static touchdown.

[Cambridge University | CUED | Structures Group | Geotechnical Group]

This page is maintained by rcb@eng.cam.ac.uk (last update 16 March 2005)