This research involves developing and demonstrating a physics-based predictive tool for gas particle flows. Physical phenomena involving gas particle flows are ubiquitous in engineering. Spray combustion, pollutant transport, helicopter brownout are a few such examples. The understanding and prediction of gas particle flows are critical in engineering design to enhance performance (e.g. improved combustion efficiency) or mitigate harmful effects (e.g. helicopter brownout). The primary objective of this project is to develop a tool, which can be applied to a wide range of gas particle flows, and is fast and accurate enough to impact the design process. Predictive computational models developed in this work will have direct applications in the design of automotive and aircraft engines, chemical processing plants, rocket engines, diesel and spark ignition engines, and industrial furnaces.
Several methods are popular in gas particle multiphase flows, for example, the Lagrangian method, the volume of fluid method and the level set method. Although they all achieved success in some applications, they also possess some fundamental weaknesses. In this project, the investigators apply the recently developed direct quadrature method of moments to treat the kinetic equation in an Eulerian framework. Since it is not necessary to follow and resolve each individual particle, as in the Lagrangian method, the DQMOM method is several orders faster. In addition, an unstructured grid based high-order spectral difference method is employed to discretise the governing equations to achieve higher resolution, and better accuracy and efficiency for complex configurations than the current state-of-the-art second-order methods. The developed tool is validated with several benchmark test problems.
Kinetic theory is a useful theoretical framework for developing computational fluid dynamic (CFD) models for disperse multiphase flows. For example, Lagrangian particle tracking methods such can be formulated in terms of a kinetic equation written in an Eulerian framework. For most applications, direct solution of the kinetic equation is intractable due to the high dimensionality of the phase space. A key intellectual challenge is thus to reduce the dimensionality of the problem without losing the underlying physics. Lagrangian methods and Eulerian multi-fluid models are two widely used CFD tools for accomplishing this task. In theory, starting from the same kinetic equation, Lagrangian and Eulerian CFD models should yield identical results for multiphase statistics (e.g. phase volume fractions, phase velocities, etc.) but this is often not the case. More often than not, the reason for the discrepancy can be found in the closures invoked in deriving the Eulerian CFD model. In this project, we have developed a more general closure approximation based on reconstructing the distribution function in the kinetic equation from its moments using a quadrature-based methodology. In principle, this Eulerian CFD approach can treat polydisperse multiphase flows as accurately as the corresponding Lagrangian approach. Using representative examples from liquid sprays, gas-particle flow, and bubbly flow, we have developed and demonstrated the underlying fundamental numerical algorithms needed to use quadrature-based moment methods for simulating polydisperse multiphase flows.
