부산대학교 도서관

This thesis describes the design and commissioning of an experimental test rig to characterise and measure buoyancy-induced heat transfer in rotating cavities, simulating an axial compressor in an aero-engine. The purpose of the rig is to provide experimental data to validate computational/theoretical models and thermo-mechanical codes used by engine designers. The work is principally experimental in nature with a strong emphasis on developing appropriate instrumentation for reliable heat transfer measurements performed in a rotating environment at high levels of mechanical stress. Much work is also generated in terms of data analysis, theoretical modelling and finite element analysis. It is now urgent to reduce harmful emissions at altitude to avoid catastrophic climate change. The next generation of aero-engines (e.g. ultra-high-bypass ratio and geared turbofans) have technology drivers demanding smaller core dimensions; these require shorter compressor blades with increased sensitivity to clearance losses and reliability across a flight cycle. To enable new technologies and architectures, engine designers must predict the temperature and radial growth of compressor disks (to which the blades are attached), which affect the blade tip clearance and compressor stability. These design issues are particularly acute during engine transients. Efficient operation with reduced fuel consumption is dependent on highly efficient engine components, particularly compressors, leading to the requirement of secondary air systems (SAS) for cooling purposes. Aero-engine compressors typically feature co-rotating disks which form cavities bounded by a shroud, and with an axial through-flow of air. Heat is transferred to the shrouds and disks from the hot, compressed air in the main gas path. If the shroud is hotter than the axial through-flow, buoyancy-induced flow can occur in the open rotating cavities. The radial distribution of temperature and thermal stress in the disks governs the expansion of the rotor and the running clearance between the compressor blades and the outer casing in the mainstream cavity. The issue is also pertinent to industrial gas turbines. Buoyancy-induced flow is a strongly conjugate problem: the temperature distribution on the disks affects the flow in the cavity, and vice-versa. As a preliminary study, the thesis describes a numerical disk-growth model for a single, isolated rotating disk, assessing the impact of imposed generic radial distributions of temperature. The impact for future engine operational conditions with shorter compressor blades and higher engine pressure ratios is determined. The first main portion of the thesis describes a thermo-mechanical design of the rig including a titanium disk pack (including thermocouple instrumentation) to ensure operation at 8,000 rpm and 150 oC, with a design life of 10,000 cycles. The design aspects include novel thermocouple grooves to keep stress levels in the elastic range. Key features of the rig include modularity of components and access to the test section. The thermo-mechanical design was produced as an internal report, reviewed and approved by engineers at Rolls-Royce. The second main portion of the thesis describes the collection and analysis of experimental data over a wide range of non-dimensional operating conditions. The experiments have helped characterise key features of buoyancy-induced rotating flow relevant to engine designers and for the validation of numerical and theoretical models. The work in this thesis has led to four peer-reviewed journal publications, three of which were presented at annual Turbo Expo for the American Society of Mechanical Engineers.