Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.777691
Title: Modelling thin films and droplets using smoothed particle hydrodynamics
Author: Ehigiamusoe, Nowoghomwenma Noel
ISNI:       0000 0004 7963 4654
Awarding Body: Heriot-Watt University
Current Institution: Heriot-Watt University
Date of Award: 2017
Availability of Full Text:
Access from EThOS:
Access from Institution:
Abstract:
The small scale and length of time, in addition to the exciting phenomena (such as the inner processes within the fluid) presented by droplets and thin films, make experimen tal observations difficult and, thus require the need for computational models capable of reproducing these processes in engineering. Current computational models are dominated by Finite Element and Finite Volume methods; whilst this has advanced to a high level of improvement and understanding, they however, lack the capacity to capture large deformations adequately and applied on complex systems, which is mainly due to its dependence on their mesh requirements. The present research pro posed and developed a new method of solving thin films and droplet problems using a full Lagrangian approach known as Smoothed Particle Hydrodynamics (SPH). SPH solves the continuum set of conservation equations and provides the ability to accu rately track the fluid or material history throughout its lifetime. The thesis explores and develops new and novel single phase SPH models to reliably treat and handle the dynamic nature of surface tension effects over long simulation time scales. In particu lar, Intermolecular Interaction Force (IIF), Continuum Surface Force (CSF), Contact Line Force (CLF) and Disjoining Pressure (DP) models are developed and applied on a variety of surface tension dominated flow problems and the results, where possi ble, are validated against known analytical and experimental findings, which include investigations of droplet oscillation, wetting on substrate, contact angle hysteresis and thin film rivulet flows to highlight the capability of the proposed developed SPH methodology and models. The SPH solver is developed from scratch using C++ to maximise extensibility of the methodology and computational performance.
Supervisor: Lee, Yeaw-Chu Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.777691  DOI: Not available
Share: