DEVELOPMENT OF ADVANCED TURBULENCE MODELS FOR BETTER PREDICTION OF TURBULENT FLOWS

Abstract

Turbulence remains one of the most challenging phenomena in fluid mechanics, characterized by chaotic, multiscale, and three dimensional fluctuations in velocity, pressure, and vorticity. Accurate prediction of turbulent flows is crucial across engineering applications, from aerospace and automotive design to energy systems and environmental modeling. Conventional turbulence modeling approaches, such as Reynolds Averaged Navier–Stokes (RANS) and Large Eddy Simulation (LES), have made significant contributions to computational fluid dynamics (CFD), but limitations persist in capturing complex flow separation, anisotropy, and transitional regimes. This study focuses on the development of advanced turbulence models that integrate higher order closure schemes, non local pressure–strain formulations, and hybrid RANS–LES frameworks. The proposed models aim to enhance predictive accuracy in high Reynolds number flows while maintaining computational efficiency. Numerical experiments are conducted on canonical test cases—including turbulent boundary layers, backward facing step flows, and bluff body wakes—to evaluate model performance against experimental data and Direct Numerical Simulation (DNS) benchmarks. Results demonstrate notable improvements in the prediction of mean velocity profiles, turbulence intensities, and separation/reattachment points compared to baseline models. The findings highlight the potential of these advanced turbulence models to bridge the gap between computational feasibility and physical accuracy, providing a robust framework for the next generation of CFD tools in industrial and scientific applications.