Dr Cihan Ates and Abdallah Alshanawani presented their work "Towards Data-Driven Design of Flow Blurring Atomizer" at the COMSOL 2023 conference in Munich. Their presentation explores the synergy between machine learning, experimentation and CFD and how these components can be used to uncover the momentum transfer mechanism underlying atomization.

The atomization of liquids is based on various forces that disturb the surface of the liquid. In flow-blurring (FB) atomization, turbulent structures are created inside the liquid channel to achieve this effect. The aim of the study is to create a comprehensive database of transient gaseous coherent structures under different operating conditions and nozzle designs to shed light on the underlying momentum transfer mechanism. The collected data are then processed using machine learning to explore the optimal sensor configuration to capture informative signals and determine the feasibility of estimating surface shear stress by simple pressure measurements in a real setup. The analysis shows that it is not only possible to characterise shifts in the flow regimes with a single pressure sensor, but also to estimate the surface shear stress acting on it. For more details, you can check our proceeding paper.

The results have practical implications for the op