Identifcation of spudcan fixity for a jack-up rig

Abstract

To estimate soil stiffness parameters under spudcan foundations and assess the performance of a jack-up rig, a strategy for identifying spudcan fixity of a jack-up rig based on the measured acceleration responses is developed. The finite element model of a jack-up rig is set up, in which the interaction of soil and a spudcan foundation is modeled as a system of horizontal, vertical, and rotational elastic springs. The 'measured' acceleration responses along each leg is archived by simulating the jack-up under a selected storm loading in the North sea, including random wave, current and wind loadings. The random wave loading is calculated using JONSWAP water surface elevation spectrum in conjunction with Morison's equation. The spudcan fixity of a jack-up rig is effectively identified using an improved genetic algorithm (GA) method. The identification strategy proposed works by matching the time histories of the measured accelerations up with those of the numerical simulated accelerations through an objective function in GA. The results indicate that the soil stiffness parameters under spudcan foundations are exactly identified in noise free case. Error less than 1.2 % in identified rotational and vertical spring stiffnesses of interest is archived with the worst case of 10 % noise and only two acceleration measurements at hull. In addition, it is observed that the vertical and rotational stiffness parameters are more accurately identified than the horizontal one. The identified stiffness values of spudcan fixity help to enhance the understanding of jack-up foundation behavior and improve the operational envelop for a specific jack-up design. With full-scale measurements of environmental conditions and dynamic responses of a jack-up rig, this strategy can be employed to verify recent findings on the conservatism of predicting soil stiffness parameters. Copyright © 2009 by ASME.