Lattice Boltzmann Methods for Real Time Simulations of Complex Physics

点击翻译按钮获取中文摘要

Complex physics and multi-phase flows with/without heat and mass transfer can be encountered in many industrial, biological, environmental, and geophysical processes; for instance, blood flow in deformable channels. The blood consists of red and white blood cells and platelets. These immiscible particles are deformable and float in a plasma which flows in elastic channels, adding further complexity to the problem. In industry, most flows are turbulent and usually associated with multi-physics. Examples are the transport of bubbles and droplets in water and oil separation processes. Understanding the physics and the effect of the varying parameters underlying such problems are of vital interest to industry and academia. Recently, there has been a growing interest in applying the Lattice Boltzmann method (LBM) to discrete mechanics which has emerged as a powerful technique to tackle such physical problems. The LBM is easy to code, and enjoys the flexibility of allowing the inclusion of multi-physics to solve the problems at hand, adding an array of further possible applications. However, the method is relatively new, much room remains for expanding it especially in the realm of multi-physics. The turbulent flow model based on the LBM is also in great need of further exploration, as are multi-scale problems. The main objectives of this proposal are: to understand the physics of complex flows with deformable boundaries for laminar and turbulent flows with heat and mass transfer, and to develop mathematical models for such systems using discrete method based on the LBM. The developed models will be applied to specific problems. Also, the model will take advantage of the local nature of LBM to utilize multi-processor graphic card to speed real time simulations. The long term objective is to develop a unified platform to model fluid flow and solid interaction for a wide range of the controlling parameters. Moreover, lab scale experiments will be performed to validate the developed models. The experiments and flow visualizations will add to reinforce our understanding of the physics of the problem. The project will train several graduate students for solving multi-disciplinary, challenging industrial/biological problems.