Axial compressor blade design for desensitization of aerodynamic performance and stability to tip clearance.

摘要

Tip clearance flow is the flow through the clearance between the rotor blade tip and the shroud of a turbomachine, such as compressors and turbines. This flow is driven by the pressure difference across the blade (aerodynamic loading) in the tip region and is a major source of loss in performance and aerodynamic stability in axial compressors of modern aircraft engines. An increase in tip clearance, either temporary due to differential radial expansion between the blade and the shroud during transient operation or permanent due to engine wear or manufacturing tolerances on small blades, increases tip clearance flow and results in higher fuel consumption and higher risk of engine surge. A compressor design that can reduce the sensitivity of its performance and aerodynamic stability to tip clearance increase would have a major impact on short and long-term engine performance and operating envelope. While much research has been carried out on improving nominal compressor performance, little had been done on desensitization to tip clearance increase beyond isolated observations that certain blade designs such as forward chordwise sweep, seem to be less sensitive to tip clearance size increase. The current project aims to identify through a computational study the flow features and associated mechanisms that reduces sensitivity of axial compressor rotors to tip clearance size and propose blade design strategies that can exploit these results.;The methodology starts with the design of a reference conventional axial compressor rotor followed by a parametric study with variations of this reference design through modification of the camber line and of the stacking line of blade profiles along the span. It is noted that a simple desensitization method would be to reduce the aerodynamic loading of the blade tip which would reduce the tip clearance flow and its proportional contribution to performance loss. However, with the larger part of the work on the flow done in this region, this approach would entail a nominal performance penalty. Therefore, the chosen rotor design philosophy aims to keep the spanwise loading constant to avoid trading performance for desensitization. The rotor designs that resulted from this exercise are simulated in ANSYS CFX at different tip clearance sizes. The change in their performance with respect to tip clearance size (sensitivity) is compared both on an integral level in terms of pressure ratio and adiabatic efficiency, as well as on a detailed level in terms of aerodynamic losses and blockage associated with tip clearance flow. The sensitivity of aerodynamic stability is evaluated either directly through the simulations of the rotor characteristics up to the stall point (expensive in time and resources) for a few designs or indirectly through the position of the interface between the incoming and tip clearance flow with respect to the rotor leading edge plane. The latter approach is based on a generally observed stall criteria in modern axial compressors. The rotor designs are then assessed according to their sensitivity in comparison to that of the reference rotor design to detect features that can explain the trend in sensitivity to tip clearance size. These features can then be validated and the associated flow mechanisms explained through numerical simulations and modelling.;Analysis of the database from the rotor parametric study shows that the observed trend in sensitivity cannot be explained by the shifting of the aerodynamic loading along the blade chord, as initially hypothesized based on the literature review. Instead, two flow features are found to reduce sensitivity of performance and stability to tip clearance, namely an increase in incoming meridional momentum in the tip region and a reduction/elimination of double leakage flow. Double leakage flow is the flow that exits the tip clearance of one blade and proceeds into the clearance of the adjacent blade rather than convecting downstream out of the local blade passage. These flow features are isolated and validated based on the reference rotor design through changes in the inlet total pressure condition to alter incoming flow momentum and blade number count to change double leakage rate. In terms of flow mechanism, double leakage is shown to be detrimental to performance and stability, and its proportional increase with tip clearance size explains the sensitivity increase in the presence of double leakage and, conversely, the desensitization effect of reducing or eliminating double leakage. The increase in incoming meridional momentum in the tip region reduces sensitivity to tip clearance through its reduction of double leakage as well as through improved mixing with tip clearance flow, as demonstrated by an analytical model without double leakage flow.;The above results imply that any blade design strategy that exploits the two desensitizing flow features would reduce the performance and stability sensitivity to tip clearance size. The increase of the incoming meridional momentum can be achieved through forward chordwise sweep of the blade. The reduction of double leakage without changing blade pitch can be obtained by decreasing the blade stagger angle in the tip region. Examples of blade designs associated with these strategies are shown through CFX simulations to be successful in reducing sensitivity to tip clearance size.