Unsteady fluid flow around certain bluff bodies.
PolpitiyeSisira J.
2015
It is shown in this thesis that fluid dynamic forces on unsteadily moving bluff bodies depend on the history of motion as much as on the velocity and acceleration of motion. An empirical relationship between the motion of the body and the resulting force is obtained by analysing the effect of the history of motion on the fluid dynamic force at any instant. The fluid dynamic force, velocity and acceleration are obtained as functions of time, by oscillating test models in water while they are being towed at constant speed. The test models used are: 1. a two-dimensional circular cylinder, 2. a rectangular block with square frontal area and fineness ratio of 3:1, 3. a cruciform parachute canopy with arm ratio of 4:1, and 4. a ring-slot parachute canopy. The functions by which the history of flow affects the future forces, are evaluated by using the Convolution Integral. The results show that the effects due to history of both velocity and acceleration are by no means negligible, that is the velocity and the acceleration at a specific time prior to any instant is so domineering that the fluid dynamic force can approximately be expressed as being delayed by this period of time. This 'time-delay', or time lag (as opposed to phase-lag) in the part of the measured force is found to be independent of the frequency of excitation. In the light of this evidence, a prediction model is suggested for estimating unsteady fluid forces. The data required for the application of this prediction model are obtained experimentally. Chapter One of this thesis gives a brief explanation of the historical background of unsteady fluid dynamics. The effects of acceleration on the fluid dynamic force, in both ideal and real fluids, are discussed in Chapter Two. Explained in Chapter Three are the techniques used for building the force prediction model, and data acquisition. The experimental procedure is explained in Chapter Four. Chapter Five gives the empirical form of the prediction model, and some data that are used in association with this model.