Impact of Axial Casing Groove Geometry on the evolution of flow structures and turbulence in an axial compressor

Mitigating surge and rotating stall in axial turbomachines has been the focus of many studies over the years. It is widely believed that these adverse phenomena are primarily associated with tip leakage flows, hence endwall casing grooves have shown promise in extending the stall margin. This experimental study uses stereo-PIV measurements to characterize the interactions of different axial casing grooves (ACGs) with the flow in the tip region of an axial turbomachine. The tests involve grooves with the same inlet overlapping with the rotor blade leading edge, but with different exit directions located upstream. The first part of the thesis elucidates the flow structures associated with the different grooves and identified their impact on the performance. A better understanding of groove-passage flow interactions paves way for better groove design capable of both delaying stall and improving efficiency. In the next section, the impact of the different ACGs on the structure of turbulence is discussed. Interaction of the tip flow with the ACGs modifies the highly anisotropic and inhomogeneous passage turbulence. Most TKE distribution trends can be explained by the local turbulence production rates, elucidating the different mechanisms involved and providing a unique database for turbulence modeling. Finally, the evolution of flow structures and turbulence in the entire passage across multiple blade rows, both with untreated endwall and semicircular grooves installed, are discussed for different flow regimes. For this, stereo-PIV measurements are performed in multiple axial planes covering entire passages, from the hub to the tip, both between and within blade rows. Turbulence is both spatially inhomogeneous and anisotropic in the entire passage, with each stress component showing different trends. Distributions of eddy viscosity measured by dividing the deviatoric stresses by their corresponding mean strain rate components show substantial variation in magnitude and sign across the passage, showing very different trends among components. Anisotropy Invariant Maps (AIMs) highlight the evolution of the spatially varying structure of anisotropy across the machine. Variations in the alignment of the Reynolds stress eigenvectors are also introduced and discussed. Such data are essential for the development of machine learning based Reynolds stress models for turbomachinery flows