INTRODUCTION

Pronounced variability has always been present in forces measured on cylinders immersed in waves The variability carries through to wave-force coefficients derived from the force measurements by relating them to the kinematics of the undisturbed wave train It has now long been realised that this variability is not just some quirk of the measurements or analysis but, rather, an inherent feature of the fluid loading on marine structures (see, for example, Pearcey and Bishop, 1979)

The purpose of this paper is to examine the hypothesis that the variability is, at least in part, associated with the fact that the fluid impinging on the cylinder at any given instant (the incident flow) has been embedded in the wake in an earlier phase of the cyclical motion This wake re-encounter will introduce strong perturbations to the incident kinematics and hence to the forces on the cylinder Wakes by nature are highly turbulent and so the perturbations in kinematics and forces will contain a distinct variability They are also likely to cause differences between the forces encountered in different circumstances, for examples between regular waves and irregular seas

Few, if any, direct measurements have been made of the perturbations to incident kinematics But evidence can be inferred from experimental observations of the resulting perturbations in the flow on the cylinder itself Such links between perturbations in the flow past the cylinder and those in the incident kinematics are perhaps most securely established for conditions which are idealisations of those existing for cylinders in waves

It is in this context that the paper re-examines some flow observations presented by Pearcey, Matten and Singh (1986) The cylinder is first considered in simple harmonic, planar oscillatory motion relative to the fluid, this generates many of the essential features of a vertical cylinder in regular, unidirectional waves Reference is made specifically to experiments In which the cylinder was oscillated about a stationary mean position in otherwise still water The idealisation is then extended to represent the effect of a superimposed current by applying a uniform translation to the oscillating cylinder

A further example is drawn from measurements on a cylinder moving in a near circular orbit in otherwise still water (Chaplin 1988)

1 The Sign of the Wake-induced Perturbations to Velocity

One might rat first sight expect the wake of a circular cylinder to be a broad region of average viscous retardation, (¿Uo)t say, (Fig l(a)), from the velocity (UoIt of the unndisturbed incident stream So, in oscillatory flow, the incident velocity for a wake re-encounter would be (Uo)t2 - (¿Uo)t1

But, it is also known that the wake frequently contains discrete vortices, often in pairs of opposite sign which can induce local velocity increments, (~Uo)t (Fig l(b)) The incident velocity for a wake re-encounter would then be ((Uo)t2 + (~Uo)t1

Figure 1 Velocity perturbations in the wake of a cylinder in a uniform stream of velocity Uo, (a) In the absence of vortex shedding, (b) with vort ex shedding (Available in full paper)

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