Title: Adaptive Multi-Scale Simulations of Liquid-Gas Flows for Accurate Spray Predictions
Liquid atomization is critical to many engineering applications such as fuel combustion and powder production. These flows are characterized by a multi-scale breakup process: large-scale instabilities grow on liquid jets, leading to the development of thin films and filaments which then break up into a fine spray of droplets. Fully capturing this multi-scale breakup with traditional CFD methods is challenging and computationally expensive. Without sufficient grid resolution, the final breakup process is dominated by grid-dependent numerical errors, resulting in erroneous spray predictions.
In this talk, an efficient multi-scale Volume of Fluid framework is presented. This method is enabled by four main components. First, a novel adaptive liquid-gas interface reconstruction captures fluid features from large length scales to the sub-grid scale. Second, on-the-fly feature classification allows for the selection of locally optimal interface geometries (e.g., films, filaments, droplets) during simulations. Third, physics-based breakup models convert thin liquid structures into sub-grid droplets. Fourth, adaptive grid refinement facilitates simulations with the large domain sizes and higher refinement levels necessary for industry-relevant applications. This full multi-scale method is applied to spray-atomization test cases, and agreement with experimental images and drop size data is shown.
Dr. Cahaly's Linked-in page: https://www.linkedin.com/in/andrew-cahaly-70bba7204/