Experimental studies of supersonic compressible flows in Overexpanded nozzles have
proved the existence and interaction of several physical phenomena: Supersonic jet, shock waves, boundary layer separation, reversed flow, viscous and turbulent mixing layer. These complex phenomena can significantly affect the performance and reliability of propulsive nozzles. This work focuses on the physical analysis and numerical simulation of turbulent separated flow in supersonic nozzles, operating in under-overexpanded conditions. The turbulence is modelled using a statistical approach (URANS) on a generalized coordinates, using the two-equation model of transport (SST Menter). The system of equations governing the flow is solved using the finite volume method in structured grid. The time integration is performed by the fully implicit numerical scheme of predictor-corrector type of Mac-Cormack. While the convective flows are discretized with shock capturing schemes (Roe, Steger-Warming). Viscous terms are discretized with a second order centred scheme. The Obtained numerical results have detected the various phenomena observed experimentally.
Experimental studies of supersonic compressible flows in Overexpanded nozzles have
proved the existence and interaction of several physical phenomena: Supersonic jet, shock waves, boundary layer separation, reversed flow, viscous and turbulent mixing layer. These complex phenomena can significantly affect the performance and reliability of propulsive nozzles. This work focuses on the physical analysis and numerical simulation of turbulent separated flow in supersonic nozzles, operating in under-overexpanded conditions. The turbulence is modelled using a statistical approach (URANS) on a generalized coordinates, using the two-equation model of transport (SST Menter). The system of equations governing the flow is solved using the finite volume method in structured grid. The time integration is performed by the fully implicit numerical scheme of predictor-corrector type of Mac-Cormack. While the convective flows are discretized with shock capturing schemes (Roe, Steger-Warming). Viscous terms are discretized with a second order centred scheme. The Obtained numerical results have detected the various phenomena observed experimentally.
