MIXING MECHANISM OF MULTI-SCALE FLOW IN TIP REGION OF TURBINE ROTOR

The tip leakage loss of freestanding turbine rotor with transonic flow conditions is one of the important sources of the internal loss of high-pressure turbine. It mainly includes the internal loss of the tip and the mixing loss of the leakage flow and the main flow. The latter is the main contributor of the tip leakage loss. The mixing between the leakage flow and mainstream flow is a complex physical process, which contains abundant multi-scale flow structures. Therefore, it is of great significance to understand the mixing mechanism of multi-scale flow in order to accurately evaluate the tip leakage loss and improve the aerodynamic performance of the turbine. In the present, the high-precision flow field in the tip region is obtained by Detached Eddy Simulation, and the multi-scale flow are decomposed by Kolmogorov Spectrum Consistent Optimization (KoSCO). Furthermore, the mixing coefficient is defined by combining Lagrange and Euler method. The contribution and physical mechanism of the local multi-scale flow to the mixing is studied carefully. Finally, the mixing coefficient is associated with the velocity field. The results show that the KoSCO method in this paper can effectively induce the multi-scale flow structure in the tip region of turbine rotor. The mixing coefficient defined in this paper can effectively evaluate the difference of the contribution of different scale flows to mixing. The contribution of different scale flows to mixing is scale dependent. The mixing coefficient decreases with the decrease of flow scale. However, when the flow scale is smaller than a specific value, the mixing coefficient is almost the same. The physical mechanism of flow mixing of different scale flow is quite different. The transport and diffusion dominate mixing process in large-scale flow and small-scale flow, respectively. Specially, both transport and diffusion impact the mixing process. The relationship between the mixing coefficient and the velocity gradient is found, and the correlation formula between the mixing coefficient and the velocity field gradient is established to evaluate the mixing strength.