Rahul Radhakrishnan
Computational Fluid Dynamics Laboratory, School of Mechanical Engineering, Vellore Institute of Technology, Vellore-632014, India
Sekarapandian Natarajan
Computational Fluid Dynamics Laboratory, School of Mechanical Engineering, Vellore Institute of Technology, Vellore-632014, India
Ashok Kannaiyan
Computational Fluid Dynamics Laboratory, School of Mechanical Engineering,
Vellore Institute of Technology, Vellore-632014, India
Mehmet Karaca
ICARE-CNRS 1c, ave. de la Recherche Scientifique F-45071 Orleans cedex 2, France; Department of Aerospace Engineering, Faculty of Engineering, Middle East Technical University, Ankara, Turkey
Jayaraman Kandasamy
Department of Aerospace Engineering, Faculty of Engineering, Middle East Technical University, 06800 City, Çankaya Ankara, Turkey
In this research paper we investigate, using numerical
simulation, film cooling of a horizontal flat plate with its
top surface exposed to hot mainstream air. We achieve film
cooling by passing a coolant air in the cross-stream direction
through a row of cylindrical film cooling holes drilled across
the thickness of the flat plate. A parametric study on the
complex interaction of film cooling jet and mainstream is
critical for understanding the film cooling process, especially
in the high free-stream turbulence conditions. To fulfil this
objective, we develop a validated 3D virtual simulation
framework using ANSYS Fluent. Realizable k − ε (RKE), SST k − ω, BSL k − ω, and Reynold Stress Model (RSM) RANS models are used to predict the effects of turbulence. To optimize the computational cost of the simulation, we reduce the computational domain size by the virtue of symmetry in the geometrical and flow conditions and by modeling the coolant flow without the plenum. We validate the developed numerical model against experimental data available in the literature for Turbulence intensity (Tu) values of 0.5% and 20%, momentum flux (I) and density ratios of 0.156 and 1.05, respectively. In the validation studies, we compare the obtained span-wise and stream-wise numerical thermal field data and report the discrepancies in the predictive capability of the chosen turbulence models. Finally, we draw a few conclusions on the merits and demerits of these models towards the accurate simulation of this application.