Identification of the nonlinear roll damping and restoring moment of a FPSO using Hilbert transform

Roll motion is a major concern of ship and offshore operators because it has a direct impact on the boarding comfort of crews, downtime of the process plant and structural integrity of the hull and appurtenance system. Nevertheless, the nonlinear effects involved in the roll motion of ship and offshore structures are not only incompletely understood, but also inaccurately predicted during the design stage. Therefore, efforts have been made to identify the dynamic characteristics of the roll motion of a FPSO using a novel dynamic system identification scheme based upon the Hilbert transform. First, the system identification scheme using the Hilbert transform was applied to the idealized single DOF nonlinear oscillator problem with quadratic damping and quintic stiffness terms. The solution of the single DOF problem was solved numerically using the 4th order Runge-Kutta method and the target damping and stiffness coefficients were determined by applying the Hilbert transform to the numerically obtained free decay signal. A comparison was also made with traditional logarithmic decrement technique for the damping coefficient. Second, the system identification method was applied to the model test results of a FPSO with a bilge keel, where the nonlinear effect of roll damping and restoring is expected to be dominant. This methodology can be applied to full scale measurement data to achieve a clearer understanding on the nonlinear effect of roll motion.