ABSTRACT

This paper reports on some of the results of an experimental test program on the dynamic response and interactive behavior of marine risers and tendons, which was conducted in the deep water basin at the Offshore Technology Research Center (OTRC), Texas A&M University. Specifically, the paper focuses on the nonlinear system analysis of the interactive behavior of a pair of long tensioned cylinders in close proximity exposed to unidirectional random wave loading. The Volterra series approach was utilized, since it is a nonparametric method and thus does not require a priori information about the unknown physical system to be modeled. By using well developed high-order statistical signal processing techniques, the Volterra kernels can be determined by calculating the various higher-order moments of the excitation-response measurements. Time-series data of wave elevations, displacements, and reaction forces for a pair of cylinders spaced 3.5 diameters apart and in tandem array with respect to the wave flow were organized as inputs-outputs of unknown systems and were modeled using nonlinear third-order Volterra models. Two different pretension levels of the cylinders were investigated, namely 9.815 MN (.low. pretension) and 13.8 MN (.high. pretension). In the high pretension case the cylinders can be considered as responding as a cluster, with comparable displacements and reaction forces, which are strongly linearly related, particularly over the high wave energy frequency band. In the low pretension case, the reaction forces for the two cylinders show some marked differences and are not strongly linearly related. Nevertheless, the third-order Volterra model can still capture 80–90% of the observed power of the downstream inline reaction force over all frequencies.

INTRODUCTION

The design of tendon and riser systems for offshore structures deployed in deep water, such as semi-submersibles, TLP and spar platforms requires serious attention.

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