ABSTRACT The data for this study comes from a series of large-scale tests conducted in the Deka flume of the Delft Hydraulics Laboratory in Netherlands (1993). In these tests two cylinders of diameters 210 and 500 mm have been exposed to both regular and random waves, with and without currents, where the currents have been simulated by towing the cylinder along the flume at a steady speed by means of a carriage. Measurements have been made with surface conditions corresponding to both smooth and rough conditions, with cylinders in both vertical and horizontal orientations. This paper is devoted to the stochastic (nondeterministic) analysis (as opposed to the deterministic wave-by-wave analysis) of the random wave load data on the two cylinders in the vertical orientation. Therefore, the results of this study are relevant to the nondeterministic analysis and design of offshore structures, which are inevitably exposed to random wave and current loading in the ocean environment.
INTRODUCTION The most-widely accepted approach to the problem of predicting wave load on a submerged cylinder with a diameter (D) much smaller than the wavelength (L), D/L < 1/5, is due to Morison et al (1950). The drag and inertia coeffcients of this equation cannot be predicted theoretically and hence must be determined from experimental data. Extensive small-scale experiments have been conducted to study the mechanism of wave loading on cylindrical members and to derive optimum coefficients for wave load prediction. However, it is well established that these non-dimensional coefficients depend on both the flow regime (represented by Reynolds number, Re) and the Keulegan-Carpenter number, Kc. For design applications, experiments must be carried out in the post-critical re, me (high Reynolds numbers, say Re > 105 in wave-induced flow) with Kc values in the range 0 < Kc < 50.