Abstract
In aero-engine bearing chambers, two-phase shearing flows are difficult to predict as Computational Fluid Dynamics (CFD) RANS models tend to overestimate interfacial turbulence levels, leading to inaccuracies in the modelling of the flow. Turbulence damping methods have been developed to address this problem, such as Egorov’s correction, however, this method is mesh dependent and results differ considerably according to the choice of turbulence damping coefficient. In addition, this approach assumes a smooth interface between the air and oil phases when in reality they are wavy. In this paper, a Machine Learning method is used to inform an unsteady RANS turbulence modelling. It is trained using high fidelity quasi-DNS simulation data and used to provide an appropriate correction to the popular Wilcox’s standard RANS 𝑘 _− _𝜔 _turbulence model. The correction consists of a machine learning-predicted source term which is used to adjust the energy budget in the RANS transport equations. Demonstration of the approach is presented for a range of interfacial flow regimes.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
