An improved streamline curvature-based design approach for transonic axial-flow compressor blading

Abstract

The increasing demand to obtain more accurate turbomachinery blading performance in the design and analysis process has led to the development of higher fidelity flow field models. Despite extensive flow field information can be collected from threedimensional (3-D) Reynolds-averaged Navier-Stokes (RANS) numerical simulations; it comes at a high computational cost in terms of time and resources, particularly if a comprehensive design space is explored during optimization. In contrast, through-flow methods such as streamline curvature (SLC), provide a flow solution in minutes whilst offering acceptable results. Furthermore, if the SLC fidelity is improved, a more detailed component-blading study is expected.

For this reason, a fully-detailed transonic flow framework was implemented and validated in an existing in-house two-dimensional (2-D) SLC compressor performance to improve the performance results fidelity in transonic conditions. The improvements consist of two sections: (1) blade-profile modelling; (2) flow field solution. The bladeprofile modelling considers a 3-D blade-element-layout method to correctly model the sweep and lean angle, which determine the shock structure.

The essential part of the transonic flow framework is its solution, formed of two parts: (1) a physics-based shock-wave model to predict its structure, and associated losses; (2) and a novel choking model to define the choke level for future spanwise mass flow redistribution. To demonstrate the functionality of the full comprehensive transonicflow approach, the well-known NASA Rotor 67 compressor was used to prove that the inlet relative flow angle should be limited by the choking incidence at the required blade span locations. A 3-D RANS numerical simulation for the NASA Rotor 67 validated the transonic-flow model, showing minimum differences in the spanwise mass flow distribution for the choked off-design cases. The current improvements implemented in the 2-D SLC compressor/fan performance simulator enhance the fidelity not only in analysis mode, but also in design optimisation applications.