Mesoscopic Modelling and Simulation of Fluid Phenomena in Material Processing

Fluid flow is ubiquitous in nature and human life. In material processing, it plays an important role and is in association to many industrial manufacturing process. However, in most of the processes, fluid flow is not the sole phenomenon. Instead it is accompanying chemical reactions, phase transformations and substance diffusion. Many behaviours in materials processing under external fields such as in electric field are desirable to know. Studying fluid behaviours in such cases is a big challenge. The mesoscopic modelling, computation and simulation provides a promising solution. It has the advantages of properly describing the micro structured evolution and hydrodynamic effects as well as considering the thermodynamics and chemical details of substances. Meanwhile, it can balance the modeling resolution and computational cost. The objective of the work contained in this thesis was to understand such challenges. A mesoscopic simulation method based on the integration of dissipative particle dynamics (DPD), smoothed particle hydrodynamics (SPH) and computational thermodynamics (CT) has been developed. This method is capable of simulating the dynamical evolution of mixing/demixing under the factors of flow and diffusion simultaneously. The simulation demonstrates that the predictions on mixing or demixing are consistent with the phase diagram. Substance diffusivity and flow speed have a significant influence on the speed of demixing and material structure evolution respectively. The model has been further integrated to the electric current thermodynamics. This enables the simulation of interaction between fluid metal and inclusions under electric current. The simulation results show the disturbance of fluid particle velocities caused by inclusions and the trace of inclusions influenced by fluid flow. Both of the developed models are generic. The simulations can be conducted based on real parameters. Other factors such as temperature can be incorporated into the model to simulate more complicated situations. An integration of electric current effect, diffusion, thermodynamics and hydrodynamics can be developed further based on the framework provided by this work