Conceptual aero-cooling designs for high-pressure turbines in jet engines

Abstract

This thesis starts off by providing a thorough literature review of state-of-the-art existing high-pressure turbine architectures and other theoretical cooling technologies whose working principles have only been investigated outside the context of application to the high-pressure turbine. Following the compilation of a substantial amount of information pertaining to the barriers to further gains in turbojet engine efficiency, principal engine sources of failure and their mechanisms, and existing and potentially applicable thermal stress ameliorating technologies, a meticulous validation process is performed to create a robust CFD framework capable of reliably simulating the effects of various cooling technologies within a rotor-stator cavity. An analytical approach has been taken to lay a solid foundation on which the validation process is built. The computed flow and heat transfer characteristics of a simple rotor-stator cavity are compared with previous experimental and numerical studies. The results are used to interpret the heat-transfer patterns observed in the validation. A simple verification procedure for the structural analysis follows. Several proof-of-concept designs for minimising thermal gradients during flight operation have been modelled and compared following the encouraging assessment of the impact of heat transfer modelling on high-pressure turbine discs via the validation and verification processes. A number of geometry-based modifications that would require significant changes to the disc and/or system structure are first proposed and scrutinised. Upon the establishment of a combination of a TWD with concentric rib flow turbulators at the bore as an adequate design to continue developing, some alterations to the operating conditions are then perused. Although the premise of the present study lies within the conceptual design stage, the adopted approach is not unlike a typical TRL method used in industry, where once any major geometrical modifications are finalised, any additional improvements may only be carried out via altering the operating conditions. A brief pilot study to investigate potential model comparison approaches to reduce the computational resources required during the conceptual design stage follows. Initiated by exploring the possibility of establishing a relation such that further comparisons between different models could simply be made based on the temperature differences in lieu of performing a transient FEA computation every time.