Blade Profile Loss Model development for Axial-Flow Fans and Compressors performance prediction using Through-Flow Simulations

Aircraft gas-turbine engine design and analysis has represented a milestone discipline in the development of aviation. Actually the continuous demand from the civil and military sector for compactness, weight and cost reduction has pushed the research in the field towards the introduction of transonic axial-flow fans and compressors: thanks to the shock induced compression, they effectively provide high stage pressure ratio meeting at the same time the requirements of compactness and lightness. Precisely this technological necessity has driven the researchers to explore different simulation level to obtain high fidelity results and optimisation designs in acceptable timing and costs with industrial standards. In recent years the Reynolds-Averaged Navier-Stokes(RANS) Computational Fluid Dynamics(CFD) analysis has shown all its power in providing great prediction against experimental data of the flow field inside aero-engines components. However nowadays CFD is still affected by two huge problems: the computational resources and the amount of time needed. Through-flow simulation represents a valid alternative to CFD because of its capability to provide a solution of good accuracy with the need of a limited amount of computational resources and with the use of far fewer time compared to CFD providing the possibility to study complex phenomena and their effect on engine performance and efficiency in a more cost and time effective manner. The Stream Line Curvature(SLC) method is the most used numerical method for turbomachinery flow simulation in the field of through-flow tools. It assumes the flow to be two-dimensional, compressible, inviscid and steady. Because of these assumptions the SLC method has the necessity to take into account of the viscosity effects by adding some empiricism and thanks to the incorporation of the profile and shock loss models developed through the years. Thanks to these models incorporation a fully detailed analysis of an engine component can be achieved with an acceptable level of accuracy with a more effective computational and time cost. The aim of this research project is to further develop and improve the profile loss modelisation in the in-house trough flow tool developed at Cranfield University in the UTC Rolls-Royce research centre: SOCRATES (Synthesis Of Correlations for the Robust Assessment of Turbomachinery Engine Systems). In particular a new design profile loss model based only on the use of blade geometrical parameters has been developed in order to substitute the previous correlation set implemented which was based on design velocity ratios which, in reality, at the preliminary design analysis phase of the compressor stage are not still known but constitute an output of the simulation, hence the importance to avoid their use in the profile loss estimation. The new model developed constitutes the core of the profile loss estimation model that has been built by defining a set of existing models in the open literature and after having been coupled with a shock-loss model and an off??-design deviation angle model, it has been verified validated and applied producing results in good agreement with experimental data that confirm the potentiality and reliability of through-flow performance prediction and flow field analysis in the preliminary turbomachinery design.