Numerical study of particle-fluid flow in complex pipe systems
2017-02-20T23:31:24Z (GMT) by
In recent years, numerical approaches have become indispensable in studies on particle-fluid flows. In this thesis, various numerical models have been developed and applied to study the particle-fluid flows in three typical complex pipe systems which are widely encountered in many industry sectors such as energy, chemical, and mineral, to name a few. The systems considered here are: industrial scale circulating fluidized beds (CFBs), long-distance high-pressure dense-phase (LHD) pneumatic conveying systems, and novel and complex bypass pneumatic conveying systems. The numerical models used are: the two fluid model (TFM), combined model of computational fluid dynamics and discrete element method (CFD-DEM), and coarse-grained CFD-DEM model. The numerical results of the simulations are validated either qualitatively or quantitatively depending on the availability of experimental data and simulation results for comparison. The work is useful to understand, design, control and optimize the considered systems. Specifically, for industrial-scale CFBs, it is shown that the typical flow structures in the CFBs can be captured by the coarse-grained CFD-DEM approach. The particle clustering phenomenon near walls, phenomenon of solids back-mixing, and core-annular flow structure are observed. Gas-solid flow regime is a bulk behaviour resulting from the collective interactions between particles, particles and particles, particles and wall, particles and gas. Therefore, analysis of the interaction forces in conjunction with the flow behaviour of individual particles can help understand the underlying mechanisms. There is a reasonable agreement between the simulation results and experimental ones for the interaction forces. The effects of some influential variables on the gas-solid flow pattern are also studied to investigate to what extent the coarse-grained model can be used to study the large scale CFB riser and to identify the limitations and weaknesses of this model. In investigating the LHD pneumatic conveying systems, both TFM and CFD-DEM are used. Some key flow features in a stepped-pipe are captured by the developed models and analysed. The mechanisms governing the complex gas-solid flow in the stepped-pipe have been identified by the numerical results. The forces governing the motion of gas and solids have also been obtained by simulations and their importance investigated. Further, to improve the performance of pneumatic conveying by lowering the chance of pipe blockage in the stepped-pipe, a new method using inserts is proposed and tested for the first time. The effect of inserts in a conventional single pipe and a stepped pipeline has been investigated. The effects of different insert shapes on the conveying performance of a single pipe have also been studied. The results show that the conveying performance of both the single pipe and stepped pipe can be improved by implementing the inserts method. In the last part of the thesis, the CFD-DEM method is used to study the particle-fluid behaviour in a bypass system for the first time. It is shown that this model can satisfactorily capture the key features of gas-solid two-phase flow in a bypass system. The forces governing the motion of gas-solid phases have also been investigated. The simulation results are in good agreement with experimental ones. In addition, the particle-fluid flow characteristics for different geometrical conditions are also studied. The numerical results have been analysed in terms of distribution of static pressure, gas and solid velocity, gas-solid flow pattern, and interaction forces. Overall, the numerical simulation methods are proven to be vital in understanding flow mechanisms in various particle-fluid systems.