This thesis investigates simplified but representative configurations of the fan wake-OGV interaction noise, which is a major source of engine fan broadband noise during take-off and landing. To this end, Computational AeroAcoustics (CAA) simulations are performed by using the Linearised Euler Equations (LEEs) and synthetic turbulence methods. An advanced digital filter method is presented to generate divergence-free synthetic turbulence with explicit control on the resulting turbulence spectrum. The method, which is based on the Random Particle-Mesh (RPM) method and synthetic eddy methods, is able to produce two- and three-dimensional fluctuating velocity fields of homogeneous isotropic and anisotropic turbulence. It is also shown that similar levels of simulation accuracy can be achieved by using digital filter and Fourier mode methods. Nevertheless, the advanced digital filter method provides enhanced performance in terms of computational cost (up to 3:9 times faster for two-dimensional simulations in this study). CAA simulations using the advanced digital filter method are performed to improve current understanding of leading edge noise from single aerofoils. For example, the method is used to examine the distortion of turbulent structures in the leading edge region. Furthermore, a comparison between numerical and experimental noise measurements in open-jet wind tunnel experiments indicate that the advanced digital filter method is capable of reproducing experimental results with an accuracy to within 3 dB. This thesis also presents a parameter study to assess the effects of moderately anisotropic turbulence, as occurs in the fan wakes, on leading edge noise from single aerofoils. Finally, fan wake modelling assumptions, such as cyclostationary variations in turbulent kinetic energy and integral length scale, are investigated using a cascade of thin aerofoils. Results indicate that broadband noise mainly depends on the circumferentially-averaged spectrum that is perceived by the cascade, and not on the instantaneous features of the fan wakes. A parameter study on cascade noise using isotropic turbulence is also included in this thesis, where variations in the vane count, aerofoil thickness, camber, mean flow Mach number, stagger angle, and inter-vane spacing are investigated. It is confirmed that the flat plate assumption provides sufficient accuracy for the frequency range in which engine fan broadband noise is relevant.

Direct numerical simulations have been performed to study receptivity and breakdown mechanisms for hypersonic flow over blunt leading-edge configurations with imposed freestream acoustic disturbances. Both two-dimensional (2D) and three-dimensional (3D) fast and slow acoustic wave models have been used. The former has been adopted for 2D simulations over a blunt-nose wedge probe designed to measure freestream noise levels in hypersonic wind tunnels; the latter has been used to perform 3D simulations for a span-periodic blunt wedge in unswept and swept configurations, and for a three-dimensional generic forebody model. In the 2D wedge simulations, an analysis of the post-shock wave structure shows that fast acoustic waves are efficiently transmitted across the shock as refracted waves, while slow acoustic waves generate convected waves. The wall response to the fast mode highlights a resonance-modulation behaviour in the nose region. An estimation of the freestream noise levels in the DLR high-enthalpy (HEG) and low-enthalpy (RWG) hypersonic wind tunnels has been performed, showing higher noise levels for the HEG wind tunnel at high Mach numbers. The 3D wedge simulations have been used to study the characteristics of the receptivity and breakdown mechanisms associated with different wave types (fast/slow), disturbance amplitudes, and sweep angles. The fast-mode induced transition has been observed to be a much more rapid and powerful process than the slow-wave related transition, because of the role played by the fast-mode resonance mechanism at the leading edge. Finally, the numerical simulations performed for a generic forebody geometry have enabled comparison with a recent transition experiment carried out in the Mach 6 Purdue hypersonic wind tunnel in noisy conditions. In this case, slow acoustic waves show the most similar transition patterns to the experimental case, and, in particular, are more efficient than fast waves in triggering nonlinear growth of streamwise streaks, related to crossflow inflectional instabilities located in the off-centerline leading-edge region.

The focus of this study is to investigate the characteristics of a zero-pressure-gradient turbulent boundary layer in the presence of large-scale free-stream turbulence. Particular attention is given to scale interactions occurring within the turbulent boundary layer. The free-stream turbulence was generated by an active grid. The investigation was conducted as an experimental work using hot-wire anemometry and Particle Image Velocimetry. Large-scale structures occurring in the free-stream are shown to penetrate the boundary layer and increase the streamwise velocity fluctuations throughout. The near-wall peak amplitude of the streamwise velocity fluctuations are observed to increase with increasing turbulence level in the free-stream. This trend has been found analogous to high Reynolds number flows for increasing ? . Free-stream turbulence imposes an outer energy peak in the boundary layer and the energy level of this peak is observed to increase with increasing turbulence level. It is also shown that the large-scales dominating the outer region of the boundary layer have a modulating effect on the small-scales in the near wall region; this effect becomes more significant with increasing turbulence level in the free-stream. These added up to the analogy between the high Reynolds number flows and present study cases. The latest efforts in the field of high Reynolds number flow investigations are aimed towards predicting the near-wall turbulence using only the large-scale information input. Therefore, this analogy has encouraging implications towards generalising large-scale influences on the near-wall small scales. The study also presents results regarding the structural organisation inside the boundary layer using the PIV data. The coherent structures found inside the boundary layer are observed to have inclined features as consistent with the previous studies for canonical flows. The fact that the external disturbance, such as FST in this study, does not alter the organisation of the structures inside the boundary layer could potentially provide an evidence for a universal structure for all wall-bounded flows as also proposed from previous studies in the literature.

This thesis presents a numerical and experimental study of the interaction of shock waves with air-filled cavities in a liquid. At the incident shock pressures studied, this process leads to the formation of high-speed jets which propagate across the cavities, causing them to collapse. This phenomenon is modelled numerically using a front-tracking approach, which enables explicit tracking of the gas-liquid interface. In arrays of multiple cavities, this method is used to show that inter-cavity effects can lead to intensification of the collapse process under certain conditions. A number of different arrangements are presented, demonstrating that in all cases the separation distance between the cavities and the relative radius of the cavities are of key importance. This result is highlighted with the study of an array of three cavities, in which it is shown that pressures over five times higher than the maximum pressure observed in the single cavity case may be achieved. Experimentally, a single stage light-gas gun is utilised to create strong shock waves in hydrogel blocks, into which spherical cavities are cast. With this setup, high-speed imaging and optical diagnostics are used to show that cavities collapse as predicted numerically, and that light is emitted upon collapse. In the single cavity case, this light is formed initially at the position at which the high-speed jet strikes the far cavity wall, propagating out subsequently into a torus. The spectral intensity of the light emission increases consistently with increased incident shock pressure. In multiple cavity arrangements, the complex inter-cavity effects seen numerically are found to be substantiated by experimental results. The spectral intensity of the light emission is seen to vary significantly with the separation distance between the cavities, with order of magnitude increases observed for particular cases.

No description available.

A micro air vehicle (MAV) is a semiautonomous airborne vehicle which measures less than 15 cm in any dimension. It can be used to access situations too dangerous for direct human intervention, e.g., explosive devices planted in buildings and video reconnaissance and surveillance, etc. As demonstrated by flying birds and insects, flapping flight is advantageous for its superior manoeuvrability and much more aerodynamically efficient at small size than the conventional steady-state aerodynamics. Piezoelectric actuators are easy to control, have high power density and can produce high output force but usually the displacement is small. With appropriate stroke amplification mechanisms piezoelectric actuators can be used to drive the flapping wings of MAV. This research aims to develop a piezoelectric fan system with 2 degrees of freedom of motion for flapping wing MAV applications. In this project, piezoelectric fans consisting of a piezoelectric layer and an elastic metal layer were prepared by epoxy bonding. A flexible wing formed by carbon fibre reinforced plastic wing spars and polymer skin was attached to two separate piezoelectric fans to make them coupled. Two sinusoidal voltages signals of different phase were then used to drive the coupled piezoelectric fans. High speed camera photography was used to characterize the two degrees of freedom motion of the wing. Theoretical equations were derived to analyse the performance of the piezoelectric fans in both quasi-static and dynamic operations, and the calculated results agreed well with the finite element analysis (FEA) modelling results. It has been observed that the phase delay between the driving voltages applied to the coupled piezoelectric fans plays an important role in the control of the flapping v and twisting motions of the wing. Selected factors such as the gap between the two piezoelectric fans which can affect the performances of the wing have been investigated and the experimental results were compared with the FEA modelling results.

One method aircraft engine manufactures use to minimize engine cost and weight is to reduce the number of parts. A significant reduction includes reducing the turbine blade count or combining two moderately loaded turbines into one high-work turbine. The risk of High Cycle Fatigue in these configurations is increased by the additional aerodynamic forcing generated by the high blade loading and the nozzle trailing edge shocks. A lot of research has been done into the efficiency implications of supersonic shocks in these configurations. However what is less well understood is the resulting unsteady rotor forces. These unsteady aerodynamics aspects are the focus of this research. The research investigates where manufacturers might concentrate their resources to reduce Direct Operating Costs (DOC). It compares the relative financial implications of disruption events to the cost of reducing DOC by further efficiency gains. The technical aspects of the research use computational aerodynamic modelling of a high work turbine to explore the unsteady aerodynamics and the resulting rotor forces. Investigation of parametric models into the effect of reaction, axial spacing, pressure ratio, the nozzle wake profile and the significance of the rotor boundary layer in dissipating the high gradient shocks is also investigated. Data from an experimental test program was used to characterise sub- and super-critical shock boundary layer interactions to determine if they are a significant forcing function. The primary conclusions from this research include the relative merits of targeting resources into reducing disruption events rather than the relatively small financial gains which might be gained through further efficiency improvement by researching advanced technologies. The computational method is validated against an experimental dataset from a high-speed turbine stage rig. Overall, good agreement is found between the measurements and the predictions for both the detailed unsteady aerodynamics as well as the important rotor forces. The effect of different computational modelling standards is also explored. The relative significance of the primary aerodynamic forcing functions such as the nozzle wake and trailing edge shock system is evaluated. Generally the rotor forces are found to increase with lower reaction, reduced axial spacing and higher pressure ratio. However the phasing of the forcing functions is found to be a critical aspect in determining the resultant net unsteady forces. The sub-critical shock boundary layer interaction is determined to be a second order effect in relation to the other primary forcing mechanisms, however the supercritical shock boundary layer interaction is shown to be a potential contributory factor in rotor forcing. Finally, several recommendations are proposed which turbine designers should apply in the event that rotor forcing is considered to be a significant concern.

Deciding how to construct a training programme or a training exercise, and especially what and how training media and methods should be selected to deliver effective and efficient training is an ongoing endeavour that preoccupies training analysts and designers alike. There are many interactions and dependencies that one has to take into consideration when making decisions, about cost, safety, or interactions between various components of a training system (e.g. between various types of media; between media and methods; between media, methods and trainees) to produce the desired outcome. The focus of this PhD research is to develop an understanding of the challenges faced by decision-makers within the military fast-jet training domain in constructing the training and, further, to develop solutions that support the decision-making effort. A significant challenge faced by decision-makers in constructing training programmes, identified through this research, is the ever increasing amount of information that they need to have at their disposal to enable fully informed decision-making and the lack of methods and tools to facilitate the management and analysis of this information. This research specifically investigated the problem of media selection to construct the training and developed a series of concept solutions to support differentiation between training media, assessment of trainees previous experiences, management of TNA outputs, selection of instructional methods and understanding of the cognitive relationship between media, method and trainee. The thesis firstly introduces the problem to be addressed; the research context and research questions set to be answered. This research, sponsored by Engineering and Physical Science Research Council (EPSERC) and BAE Systems, is preceded by another BAE Systems funded research project (the Training Optimisation Case Study), which provided the background for the work presented in this thesis. Secondly, it reviews the literature relevant to the subject matter to understand the current state of knowledge in the area of: UK RAF training programmes construction and training media selection; assessment of competencies; impact of media and method on learning; development of decision making support systems; and construction and management of knowledge. The main part of the work presented in this thesis is the development of a series of support solutions to aid the decision-making process of construction of UK fast-jet pilot training. These include: TNA output Analysis (ToA) tool; Trainee Contextual Proficiency Profile (TCPP) tool; Training Media Classification Framework; models that map the cognitive relationship between media, method and trainee, and a unified Framework of Selection of Instructional Process alongside a novel approach towards training media selection. This research work was initially scoped through an exploratory study (a case study) into the domain area, followed by requirements elicitation. This part of research helped at identifying the issues within the problem area and in defining the research questions. The TCPP and ToA were verified through two case studies and presented alongside the rest of the research to the customer (BAE Systems) that gave positive feedback on the research outcomes.

This study describes the development and application of new analysis techniques using the structured singular value m for the clearance of flight control laws for highly augmented aircraft. Motivated by the limitations of the classical methods currently used by industry, these new techniques have been developed to provide a more rigours and efficient analysis of worst-case aircraft stability and performance characteristics. The classical approaches currently used in industry for the clearance of flight control laws are explained and evaluated. New m-tools are developed to directly address specific clearance criteria currently used by industry. Different approaches to represent uncertain linear and nonlinear systems as Linear Fractional Transformation (LFT)-based uncertainty models are described: a fast numerical approach that can easily generate LFT-based uncertainty models for nonlinear aircraft models at the expense of some conservation, a more complex symbolic approach that requires detailed information about the uncertain parameters in the aircraft dynamic equations, and a physical modelling approach which can generate LFT-based uncertainty models in a straightforward manner assuming the availability of the aircraft model in a block diagram representation. Two new algorithms for computing tight bounds on real m are introduced. Both methods are shown to be capable of generating good lower bounds on m, even for high-order uncertainty models. The application of the new m-analysis tools developed in this study to the flight control law clearance process is illustrated for a detailed fighter aircraft model called the HIRM+, for a VSTOL fighter aircraft model called the HWEM and for a civil transport aircraft model. Comparisons between the newly developed analysis techniques and the classical approaches demonstrate that m-analysis tools can significantly improve both the reliability and efficiency of the flight control law clearance process.

A robust integrated flight and propulsion control (IFPC) system is designed and analysed for an experimental vertical/short take-off and landing (V/STOL) aircraft configuration, using multivariable design techniques. The model used for the design is based on the DERA VAAC Harrier wide envelope model (WEM) airframe with a Rolls-Royce Spey engine. This provides a detailed, large-scale, interacting full-envelope, nonlinear simulation model, thought likely to be representative of the next generation of V/STOL aircraft. The centralised IFPC system is evaluated in piloted simulation trials on DERA's real-time all-vehicle simulator (RTAVS), the results of which indicate that level 1 or 2 handling qualities are achieved over the low-speed powered-lift region of the flight envelope. The application of the structured singular value, , and the -gap metric to the problem of evaluating the robustness properties of this multivariable IFPC system is presented. The centralised controller is subsequently partitioned into decentralised lower-order airframe and engine subcontrollers, in order to address implementation and testing issues. The partitioned system is seen to retain largely both the performance and robustness properties of the centralised system. Due to the particular implementation structure used for the centralised H loop shaping controller, the partitioning procedure described can be applied to general two-degree-of-freedom control systems. A scheme to guarantee maximum limits on safety-critical engine variables is developed and applied, to preserve the structural integrity of the engine during extreme manoeuvres. As the system frequently operates on the position limits of the engine actuators, an anti-windup scheme is implemented to maintain the performance of the system during actuator saturation.