Numerical investigation of a transonic nozzle guide vane under elevated loading

Abstract: Despite many new investigations over the last years, there is no indication that alternative energy conversion technologies will overtake the place of turbomachinery. Hence thermal turbines are still the most dominant movers for electricity generation.Although this leadership in the energy production does not seem to be in danger, the current drivers in turbomachinery industry are to work towards both less fuel consumption and less pollution. In order to meet the future economic and environmental goals, researchers press towards highly loaded vanes and blades. This has to be performed at maintained or improved aerodynamic performances. Increased performances and blade loading lead in turn to increased velocities and larger regions of supersonic fluid velocities and consequently general increasing of shock intensities. The biggest problem dealing with supersonic flow and high shock intensities is that the boundary layer, when walking through these regions, experiences strong pressure gradients and intense shock-boundary layer interaction. This may lead the blade to stall meaning detachment of both boundary layer and cooling-film from the wall. These effects can evidently lead to catastrophic consequences since nowadays the materials used in turbomachinery applications have temperature strengths much lower than those coming from the combustion chamber. This thanks to very complex blade and vane cooling systems.There are even other features that may take benefit from increased velocities such as an attenuation in the boundary layer growth and the static pressure distribution on the blade surface. For helping researchers studying these new geometries, a cold air annular test rig designed by “Siemens Industrial Turbomachinery AB”, it has been built and placed at “Division of Heat and Power Technology” at KTH.The present thesis has the goal to provide a numerical model for CFD calculations, optimized for boundary layer studies, able to give a good prediction of detachment of the boundary layer and losses for different working cases. A previous model was provided with a commercial software for both ideal vane and real test rig. Recover of results and adaptions of the model were performed with a new version of the same software starting from the previous model. A comparison between numerical and experimental results have shown a good match for the subsonic and transonic case. Instead, problems were met for the supersonic case. Many attempts of different boundary condition at the inlet have been run. No reliable solution has been reached with realistic pressure profile at inlet while realistic results have been found using the mass flow rate as Inlet boundary condition. At the end, an analysis of shock and detachment is provided in terms of density gradient and static entropy distribution through the blade passage. Future works may aim to solve the “supersonic problem”.