ABSTRACT

This paper addresses the practical application of a multiple-input-single-output nonlinear system identification technique on ocean structural systems. An ocean structure exhibiting nonlinear behavior due to geometric nonlinearity of mooring line angles and the complexity of hydrodynamic excitations is chosen for this analytical study. Given the input wave characteristics, wave force and the system response, the method identifies the hydrodynamic drag and inertia coefficients from the wave force model formulated by relative-motion Morison equation. The reverse multiple-input-single-output technique correctly identified the parameters of the nonlinear system. The applicability of the method is demonstrated through a numerical example with noisy periodic wave excitation.

INTRODUCTION

Complex nonlinear responses have been observed in various compliant ocean systems characterized by nonlinear mooring (restoring) force and coupled fluid-structure interaction (exciting) force (e.g. Thompson, 1983; Bishop and Virgin, 1988). Gottlieb (1991) studied the nonlinear behavior of a multi-point symmetric moored structural system under periodic excitation. Lin (1994) extended this analysis by incorporating random noise perturbations. Small body mooring systems are generally solved by a relative-motion Morison formulation (Patel, 1989). It has been observed from the literature that the hydrodynamic drag (Cd) and inertia (Cm) coefficients for sphere are not constants and reasonable estimate could be 0.1 ≤ Cd≤ 1.0 and 1.0 ≤ Cm ≤ 1.5 respectively (Grace and Casino, 1969; Grace and Zee, 1978). It is important to identity the hydrodynamic coefficients and system parameters to quantitatively examine the nonlinear behavior. Bendat (1990; 1998) has used parallel multiple-input/single-input (MI/SO) procedures for identifying parameters of nonlinear systems. A method for nonlinear system identification to determine amplitude and frequency dependent properties on different types of nonlinear systems such as Dulling, Van der Pol, etc. has been developed by Bendat et al (1992). With the input and output data known, based on multiple input/single output linear analysis of reverse dynamic system, Reverse Ml/SO technique identifies the linear and nonlinear system properties.

This content is only available via PDF.
You can access this article if you purchase or spend a download.