Improvements in the efficiency of power generation via turbomachinery are essential in order to reduce greenhouse gas emissions throughout the world. Advancements in measurement techniques are therefore crucial to understanding the main areas of energy loss in turbines and compressors. This thesis presents a novel system which allows that loss to be characterised using fast response, 3-D measurements of the pressure field within industrial scale rigs. Turbulent energy dissipation rates give an insight into where useable energy is lost from. To gain an insight into such rates, measurement techniques must be able to take data in all three dimensions simultaneously at high sampling rates, usually over 50 kHz. Traditional methods of flow characterisation such as optical techniques and pneumatic pressure probes are unable to capture the rapid fluctuations in pressure and velocity which lead to energy loss from the turbomachine. A new system was therefore designed and implemented into a 6-stage compressor rig to take fast response measurements at sampling frequencies up to 100 kHz behind the last stage stator. A fast-response 5-sensor pressure head, acquired from Kulite Semiconductor Products Inc, has been embedded into a bespoke stem to allow turbulence measurements in a range of turbomachinery applications. The five-sensor (5S) probe was calibrated for pressure sensitivity as well as aerodynamically to give total and static pressure along with velocity magnitude and direction. Individual sensors were calibrated and characterised at temperatures within a range of 200C and 500C, which corresponds to the conditions found within the final application. The probe was also used in a vortex shedding experiment where alternative eddies were detected from the 5S probe measurements in both the time and frequency domain. The aerodynamic calibration of the 5S probe consists of exposing the probe sensors to a range of flow angles in order to map their response between ±200 in both the yaw and pitch directions. This results in four non-dimensional coefficients, two to represent pressure and two to signify the flow angles. A linear interpolation method was written and implemented to deduce pressure and flow angles from experimental query points and the calibration data. The linear interpolation was used as an alternative to the standard surface fit method, where the calibration data is expressed as system of polynomial equations. It was found that the linear method was applicable to the interpolation of flow angles and gave a reduction in computation time of the order of 104. The total and static pressure values do however require the more tried and tested polynomial interpolation method due to the need for higher order interaction terms in the surface fit equation describing the terms. The fully calibrated 5S probe was then implemented into a 6-stage industrial scale rig where it acquired fast response pressure data from the flow field at the exit of the last stage vane. The data was processed to give time resolved, 3-D measurements of total and static pressure, flow angle and velocity. Due to the simultaneous capture of data from all 5 sensors, the resulting velocity vectors can be decomposed into their mean and periodic components to obtain values of energy loss from the turbomachine. The acquisition of such data from an industrial rig marks a novel advancement in the area of turbomachinery flow characterisation and the use of the 5S probe in a range of applications will begin to fulfil the need for a database of fast response data from chaotic and turbulent flow fields.

The finite element method has been used to calculate the stresses and creep deformations of flanged turbine casing models subjected to internal pressure and bolting forces. The finite element results have been compared with results from photoelastic and lead model turbine casings. An axisymmetric thin shell of revolution ring finite element has been developed to analyse casings subjected to pressure, thermal and creep loads. The thin shell of revolution ring finite element is shown to be extremely powerful and has been used to investigate the shell portions of the turbine casing away from the flange. The three-dimensional isoparametric finite elements have been used for more accurate idealisations of the turbine casing. A thick shell isoparametric finite element has also been developed which can be used with the more common hexahedral isoparametric finite elements. A solution algorithm based on a frontal technique has been developed to solve the large number of linear equations given by the finite element equations. This algorithm, which is fully automatic and uses fast access backing store, has a resolution facility which is used to recalculate subsequent creep solutions assuming that the stiffness of the structure remains constant. The creep algorithms are based on time marching techniques where the creep solutions are found for small time increments, the final solution being the sum of all the incrementa1 solutions. During each time increment the stresses are assumed to remain constant and the change in stress between time increments is kept within a preset ratio. The creep algorithms have been used to predict the creep deformation of simple structures to compare with published results. The agreement between the finite element and lead model creep results is limited. The finite clement programs have been written to be compatible with the PAFEC suite of finite element programs.

This project aims to provide an in-depth understanding on pin-fin based cooling structure for gas turbine heat transfer applications, particularly for turbine hot-path airfoil blade trailing edge cooling. Based on past researches, new cooling structures are then proposed and investigated. This thesis is comprised by four primary studies including three experimental studies and one numerical study. Transient thermochromic liquid crystal method was used to measure endwall heat transfer coefficient of the channel while lumped capacitance method was used to measure the average heat transfer coefficient of cooling structure surface in corresponding studies. The Reynolds number was evaluated and pressure drop of flow across test channel was measured by the pressure taps. Parameters, such as Nusselt number, friction factor and thermal performance index, were evaluated based on experimental results. A scaled realistic NGV (Nozzle Guide Vane) hub platform model was tested. The local heat transfer distribution of its endwall was measured by liquid crystal method. The test has been carried out at Reynolds number range from 10,000 to 40,000. Two impinging nozzle plate with nozzle diameter of 5.5mm and 11.0mm were used. General findings include low heat transfer at acute angle corner and imbalance heat transfer distribution between upstream jet impingement region and downstream pin-fin region. The heat transfer rate at pin-fin region is only 44% of that at jet impingement region. Additionally, the existence of film hole (extraction hole) upstream of pin-fin region has insignificant influence on downstream pin-fin heat transfer in this test. It is also found that the heat transfer has been enhanced by 40% when the impinging nozzle diameter was doubled. Furthermore, the buoyancy effect at inlet flow has certain impact on magnitude and distribution of heat transfer at jet impinging target surface. The new elongated pedestal structure was proposed and investigated experimentally and numerically. Four elongated pedestal test sections with D/d=5.0 and 8.0, X/d=0.8 to 1.2, S/d=1.175 to 1.5 were designed and have been tested at Reynolds number range from 6,000 to 25,000. The average heat transfer coefficient at pedestal surface has been measured by lumped capacitance method. Revealed by the results, the heat transfer coefficient of pedestal surface could be at most 70% higher than that of endwall. Meanwhile, the pedestal surface could account for 50% of overall heat transfer at specific cases. The elongated pedestal structure enhanced the endwall heat transfer up to 9 times compare to reference data. Moreover, the elongated pedestal structure achieved similar heat transfer level comparing with perforated blockage structure but obtained 3 times higher heat transfer enhancement comparing to circular pin-fin structure. Generally, the tightly spaced structure obtained higher overall heat transfer than that of widely spaced structure which is same as circular pin-fin array. Via the numerical study, the flow behavior of elongated pedestal array is more like the turning flow inside the bending duct instead of flow around pin-fin structure. An extra structure, known as split elongated pedestal, has been studied numerically. However, the split elongated pedestal did not show significant improvement as expected in heat transfer enhancement as well as overall thermal performance. Currently split opening did not lead to significant flow interaction between two split parts. But it is recommended to further investigate this structure with much smaller split opening. Furthermore, three test sections with multiply cooling structure implemented were studied at Reynolds number range from 9,000 to 30,000. In addition, the test sections were modified in order to generate non-uniform inlet flow. One key finding is that the non-uniform inlet flow generated in this study leads to 25%-30% reduction in endwall heat transfer. Compare to circular pin-fin structure, cooling structure with high duct cross-section area block ratio, such as elongated pedestal and perforated blockage, provided more desired heat transfer distribution and higher heat transfer rate. Benefited from turbulence promotion by upstream pin-fin array, the heat transfer of downstream cooling configurations have been improved by 51%, 42% and 73% for pin-fin array, elongated pedestal array and perforated blockage array, respectively.

Thermoacoustic engine (TAE) can be used to convert heat from any source into electrical energy. Despite the theoretical efficiency of the cycle is very close to the Carnot cycle efficiency but due to many practical reasons, the actual efficiency of the engine is still very low. In order to enhance the overall efficiency of the waste-heat driven thermoacoustic engine (WHTAE), it is important to understand and identify the sources of losses in the engine components as well as to suggest design modifications on some critical components in the engine. All the studies reported to date are mainly focusing on the optimisation of the regenerator and the resonator without taking into consideration some of the important issues. One common trait of all the previous optimisation efforts is that the acoustic energy dissipation through the regenerator and the loop (or bends) were not well explained. It should be noted that this study provides a more comprehensive discussion on the acoustic field and the loss mechanisms between the regenerator and the sharp bend (torus-like section) in association with the radiant heat exchanger (RHX) of a WHTAE. In this work, a simplified solution and a numerical investigation are implemented to study the convection and radiation heat transfer between the regenerator and the RHX in two of the SCORE engine configurations. Both simplified solution and numerical results reveal that bulge is about three times better in total radiation heat transfer compared to the convolution. Based on the numerical results obtained, the design of the bulge show about five times more in total radiation versus convection to the regenerator top surface. The multi microphone least square technique is employed in conjunction with impedance tube measurement method to determine the acoustic properties of the tested specimen in order to develop an experimental modelling of a TAE that works in travelling-wave condition by using absorbing materials. Eight materials and combinations are investigated to realise that using an elastic end works best for low frequency attenuation applications. The selection of the attenuation material or combination of materials should be done very carefully and is strongly dependent on the target frequency. No material can work better for all frequencies. Some of the materials are suitable for high frequency but not suitable for low frequency attenuation applications. The acoustic energy losses through the regenerator and the RHX are determined by utilising the multi-microphone travelling-wave technique. It was found that when more than 30 layers of regenerator, more flow resistance is generated, there is no significant increase in the regeneration effect. Therefore, it is unbeneficial to add more than 30 layers of mesh. Owing to the perfect contact between the working fluid (gas parcels) and the solid material, the dissipation in the regenerator is dominated by viscous losses in both ambient and hot conditions. When imposing a temperature gradient across the regenerator, the system encounters more amplification than attenuation. Straight tube has the least acoustic energy dissipation and the highest loss in acoustic energy is obtained by the convolution RHX configuration. The loss in acoustic energy for the straight tube is mainly due to the viscous losses in the regenerator while the acoustic dissipation for the RHX configuration is mainly caused by the vortices generated at the two 90 o sharp bends and the sudden change of cross-sectional area. A thermoacoustic software, DeltaEC is employed to predict the acoustic energy dissipation through the regenerator and the RHX. The numerical model is found to predict the experimental results of the acoustic energy losses accurately. The DeltaEC models can be used to help on the design of future prototypes and for better optimisation of the TAEs.