Direct Numerical Simulation, Study, and Quantification of Fractal-Grid Generated Turbulent Flows

Computer modelling of some types of fluid flows can be difficult because the algorithms used in these cases need vast amounts of processing power to calculate timely and accurate results. This thesis develops an alternative algorithm that uses a particle-based approach to model the problem, allowing it to sidestep the difficulties encountered by conventional techniques. When it is applied to a class of flows with highly complex and chaotic behaviour, the algorithm successfully and accurately reproduces data close to actual real-life values. Furthermore, this was achieved while running the algorithm on a standard desktop PC, whereas previous traditional models required a supercomputer architecture to obtain the same results.