This thesis presents the formulation of a Smoothed Particle Hydrodynamics (SPH)
model and its application to a range of engineering applications. The motivation for
this research lies in the desire to accurately model viscous and turbulent free-surface
flows, including those with complete break-up of the free-surface. At present, boundary
element modelling is typically chosen to describe free-surface flows where viscous effects are not important. The Volume of Fluid method is able to model most flow phenomena, but the representation of the free-surface is insuffcient for the most complex
flows. Current SPH models have shown aptitude for modelling such flows, but there is a noticeable
lack of validation carried out in the literature. This thesis includes a thorough investigation
into established modelling techniques, extending or developing new techniques
where necessary, in order to create a versatile and accurate SPH model for free-surface
flows. Where possible, quantitative comparisons with experimental observations have
been carried out to ensure a suitable level of accuracy has been achieved.
First, a fairly basic SPH model is constructed through testing its ability to generate
and propagate solitary waves in a numerical wave flume. This is succeeded by a
thorough investigation into solitary waves breaking on a 5° slope, through which further
developments are added to the SPH model. The full process, including overturning,
post-breaking behaviour, run-up, and the subsequent hydraulic jump are quantitatively
compared with experimental measurements.
The work carried out in this thesis shows that the SPH model can successfully capture
violent free-surface flows with large deformations from the initial surface geometry.
Validation studies demonstrate that SPH can form an important part of model testing
for engineering developments involving these types of flows.