Numerical and Experimental Investigations of the Near Wall Flow in a Multi-stage Axial-flow Compressor with Casing Treatment Using TRACE and Particle Image Velocimetry

Kurzfassung

The flow in the transonic front rotor stage of a multi-stage axial flow compressor with casing treatment is investigated with experiments and simulations. Measurements of the velocity field in the blade tip region are performed using particle image velocimetry in order to determine the flow structures in this area. Furthermore the measurements are used to evaluate the quality of corresponding time-accurate 3D CFD simulations. When comparing simulations of the compressor with and without casing treatment with equal boundary conditions, an increase in total pressure ratio and mass flow is encountered for the compressor with casing treatment. The examination of the flow in the blade tip region presented a flow dominated by the interaction of different vortical structures. The origin of the vortices in the tip region can be attributed to two effects: One type of vortices arises from the interaction of the rotor blade bow shocks with the flow around the trailing edge of the inlet guide vanes. A second type of vortices results from the interaction of the inlet guide vane wakes with the tip leakage vortex of the rotor blades. The casing treatment effects both types of vortices by modulating vortex number and strength. Moreover the casing treatment influences the trajectory and the strength of the tip leakage vortex. The effects of the casing treatment result in increased and more homogeneous axial velocities in the tip region compared to a compressor with smooth wall and hence in a higher mass flow. The examination of the near tip flow field in measurements and simulations showed that simulations are able to reproduce all flow structures which can be encountered in the experiments. Trajectories of transient phenomena like vortex chains are found in good agreement between experiment and simulation. Simulations however present smaller gradients and weaker vortices compared to the measurements. Since the vortex induced flow has a major effect on the main flow through the rotor passage, the weaker vortices in the simulations result in differences in the velocity amplitudes between simulations and measurements.