The aim of this research is to study the effect of transition location on the interaction between an oblique shock-wave and a boundary-layer. A large set of direct and large eddy simulations are performed with an in-house high-order fully-parallelised finite difference compressible Navier-Stokes solver to study the inherent instability and unsteady behaviour of laminar, transitional and turbulent interactions. The numerical simulations are compared with the experiments conducted at the Novosibirsk State University as part of the EU-FP7 TFAST project, providing a better understanding of the fundamental mechanisms of the shock-wave/boundary-layer interaction (SWBLI). As well as the characteristics of the interactions, interest is also focused on methods to control the transition location. Three distinct forcing techniques are used to obtain different transition scenarios for a laminar SWBLI at free-stream Mach number of 1.5. An oblique mode breakdown caused by forcing the most unstable eigenmodes, predicted by the local linear stability theory, is compared with a bypass-like transition due to high-amplitude free-stream acoustic disturbances. A non-thermal plasma flow actuation device is also used, however showing a low applicability to supersonic flows due to the high electric power required to trigger transition. Attention is also focused on the response of a laminar shock-induced separation bubble. For both 2D and 3D configurations, a low-frequency response is found for the first time in a laminar SWBLI, even when the separation bubble is only forced internally, therefore supporting the idea that the separated region is influenced by internal mechanisms. The SWBLI is further analysed via linear and nonlinear stability approaches, including local stability theory or parabolised stability equations based tools. The response of the separated region for increasing shock intensities is studied and the stability based tools provide satisfactory results even for largely separated boundary-layers.

Direct numerical simulation has been used to study how axisymmetric vertical flow structures evolve whilst propagating horizontally in both homogeneous fluid and in fluid with a linear stable density stratification in the vertical direction. The structures studied were initially toroidal vortex rings and impulsive jets formed from a brief, horizontal injection of fluid into a quiescent domain. Previous experimental studies have demonstrated that when these initially axisymmetric structures are allowed to evolve under the influence of stable stratification, acceleration due to buoyancy acts to suppress vertical displacement of fluid particles, eventually reducing the flow to a pair of contra-rotating, planar vortices, commonly referred to as a vortex dipole. The numerical simulations documented in this thesis demonstrate the process by which the initially axisymmetric structures are transformed into late time dipoles in a stratified fluid, with the stages of this transformation categorised both through visual changes in the flow field as well as characteristic variations in kinetic energy and buoyancy variance histories that are inaccessible to the experimental work, thus allowing the energetics and vorticity fields of these flows to be directly correlated for the first time. Additionally, it has been demonstrated that while different means of imparting horizontal momentum to the fluid through an initial solution or different profiles of momentum injection may generate distinct vorticity fields at the early time, the energetics, scaling behaviours and agreement with theoretical models appear universal across the late time dipoles formed from these cases, which has not been addressed directly in previous literature.

The closure of turbulence models at all levels of fidelity is addressed, using unconventional methods that rely on data. The purpose of the thesis is not to present new models of turbulence per se, but rather the main focus is to develop the methodologies that created them. The main tool, Gene Expression Programming, is a versatile evolutionary algorithm. Implementations of the algorithm allow for symbolic regression of scalar and tensor fields and the clustering of data sets. The last two applications are novel algorithms. Scalar field regression is used to construct length scale damping functions for Hybrid RANS/LES. Direct Numerical Simulation snapshots are filtered to mimic Hybrid RANS/LES flow fields and from this new damping functions are created. Two closures are constructed, one from data in a turbulent pipe and another from slices along the classic backward facing step geometry. The new closures are tested for a range of separated flow applications. Tests alongside existing closures of the same class show that both new methods adapt to the local mesh resolution and turbulence level at least as well as other hybrid closures. Tensor field regression is used to construct non-linear stress-strain relationships in a Reynolds-Averaged Navier-Stokes framework. A common two-equation model is modified by including a further term that accounts for extra anisotropy with respect to the Boussinesq approximation. This model term, regressed from time averaged Direct Numerical Simulation data, turns the linear closure into an Explicit Algebraic Stress Model. The training data is taken from the reverse flow region behind a backward facing step. When applied to the classic periodic hills case, the subclass of models generated are found to greatly improve the prediction with respect to the linear model. A subclass of models is created in order to test the ability of the evolutionary algorithm. The deviation from the periodic hills reference data is quantified and used as a metric for model performance. The key finding is that improved performance of the Gene Expression Programming framework corresponded to improved prediction of the periodic hills. The final application of Gene Expression Programming, the clustering of datasets, is used to group Reynolds stress structures into distinct types. Firstly, reference Direct Numerical Simulation data obtained in a turbulent channel is categorised into six distinct groups. These groups are then compared to structures from Hybrid RANS/LES. These groups help to show that Hybrid RANS/LES structures do not correctly capture the near-wall cycle of turbulence. Instead there is an artificial cycle that is characterised by an incorrect buffer layer, defined by tall, long and thin structures. Further, streaky structures lie on the interface between Reynolds-Averaged Navier-Stokes and Large Eddy Simulation. These structures are free to move in the vertical direction and seriously contribute to discrepancies in the second order statistics.

Currently, there is a growing demand to improve the aerodynamic performance of Micro-Air-Vehicles for extended mission time, higher payload capacity and improved agility. Their wings have to operate within a challenging Reynolds number regime of Re =10(4)-10(5) which is known for its low energy content in the boundary layer, causing early flow separation and loss in lift production. Flexible wings, inspired from bats, could potentially exploit given flow separations by forming lift carrying shedding structures close to the upper wing surface. The aspect-ratio is one key parameter which modifies these vortex formations and their ability to couple with the membrane. However, vortex related lift production comes at a price of increased drag and limitation in aerodynamic efficiency. Membrane wings in ground-effect could combine ground-effect related efficiency enhancement with flexibility related stall improvements. Therefore, two separate wind tunnel experiments are conducted to understand the impact of aspect-ratio and ground-effect on the fluid-structure interaction of membrane wings. Multiple high-speed recordings involve lift, drag and pitch moment measurements with a load-cell, membrane deformation measurements with photogrammetry and digital image correlation (DIC)and flow measurements with planar/stereo particle image velocimetry (PIV). Next to time-averaged quantities, reduced order models are used to group predominant flow and membrane dynamics. Synchronised fluid-membrane coupling of flexible membrane wings allows to exploit separated flow conditions to provide further lift enhancement from vortical flow formations. An exemplary membrane wing at [alpha] = 25(o) shows similar vortex-shedding to a rigid at-plate at [alpha] = 15(o), but comes with 50 % more lift production. Higher aspect-ratios are found to exploit the benefits of wing flexibility to a larger extend, showing a gain in peak-lift of up to 60% for an aspect-ratio of 2 and 31% for an aspect-ratio of 1 (in reference to rigid at-plates). Membrane wings extend their performance window in ground-effect conditions by delaying ground-effect induced premature flow separation by [DELTA alpha] = 5(o). In addition, membrane wings in ground-effect are found to be up to 30% more efficiency than rigid at-plates.

Aerodynamic optimisation using gradient-based methods has found a wide range of academic applications in the last 30 years. This framework is also becoming more and more popular in the industrial world where, most of the time, unstructured grids are largely used. In this framework, apart from the need to solve the flow field, there is the need to quickly map the aerodynamic surface in terms of some aerodynamic figure of merits such as the drag coefficient, without being limited by the computational expense related to the grid size. This is a concrete industrial need which requires the efficient computation of the grid sensitivity. A novel method based on the DGM (Delaunay Graph Mapping) mesh movement is proposed to efficiently compute the grid sensitivity required in the discrete adjoint optimisation framework. The method makes use of a one-to-one explicit algebraic mapping between the volume mesh and the solid boundary nodes. This procedure results in a straightforward computation of the gradient without the need to invert a large, sparse and stiff matrix generally associated with implicit mesh movements such as the spring or LE (Linear Elastic) analogy. The method is verified using FDs (Finite Difference) and a thorough comparison in terms of CPU time, formulation against the LE-based mesh movement and adjoint gradient is presented. The DGM-based gradient chain allows to comfortably obtain the gradient with respect to each surface mesh point. Unfortunately, these gradients cannot be used directly because of their inherent poor smoothness feature. In order to address this issue one has to use a parameterisation technique which inevitably sacrifices the design space explorablity. To bridge the gap between the free-nodes and the parameterisation approaches, a novel formulation of the CST (Class Shape Transformation) was developed and termed l-CST (local-CST). The method is based on a simple trigonometric function which works as a cut-off filter on the BPs (Bernstein Polynomials) which are used to enforce a strong on-demand local control. The method is tested on an inverse geometric fitting and its effect on the resulting aerodynamic coefficients and the pressure distribution is also analysed. The DGM-based chain allows the efficient mapping of the entire surface while the l-CST allows the combination of excellent explorablity and surface smoothness. The former is tested within the non-consistent mesh movement and sensitivity framework because there are situations where one method may be preferred over the other based on the grounds that mesh movement is a very different task than mesh sensitivity although strongly related to each other. The latter is instead tested against the free-nodes approach which offers a similar advantage in terms of discrete control although without maintaining a C2 curve unless properly smoothed.

Shock wave-boundary layer interactions (SWBLIs) are an inevitable feature of compressible flow and can have a large detrimental effect on the performance of aerodynamic applications. To address and design to accommodate them, requires detailed understanding of the underlying flow mechanisms. At present, our knowledge of these mechanisms is insufficient to accurately predict SWBLI behavior. This experimental study attempts to provide a better understanding of some of these mechanisms by focusing on the three- dimensionality inherent in oblique SWBLIs. The test configuration consists of an oblique shock wave in Mach 2 flow in a rectangular test section at flow deflection angles of 8° , 10° , and 12°. The key parameters of test section effective aspect ratio (AReff) and shock generator geometry are varied to assess their ability to amplify/attenuate the three-dimensionality of a nominally two-dimensional SWBLI. An innovative traversable shock generator with interchangeable wedge geometries allow the effects of AReff , expansion fan placement, and side-wall gap to be studied. The flow is investigated by employing Schlieren photography, surface flow visualization, static pressure measurements, Laser Doppler Anemometry and Particle Image Velocimetry. It is observed that with an increase in AReff, or a downstream movement of the expansion fan, or a decrease in side-wall gap, the SWBLI and shock-induced separation will grow. The growth of the separated region exhibits an increase in three-dimensionality and at high AReff the regular reflection is observed to evolve into a Mach reflection without an increase in incident shock strength. Models are proposed to explain the observed behavior as a function of separation growth and a reduction of the influence of free interaction theory.

In many sporting activities the athletes and their associated equipment often operate in a range of Reynolds number, Re, that is close to the critical regime where the boundary layer flow undergoes transition from a laminar to a turbulent state. It is well known from numerous studies on circular cylinder and sphere flows that boundary layer transition can be forced to occur at lower Re, and hence drag reduced by delaying flow separation, through the application of, for example, roughness to the surface. This thesis is aimed at increasing understanding of how passive flow control methods might be employed to influence boundary layer flow in order to reduce the drag of bluff bodies. A wind tunnel based research programme was undertaken to study these aspects, including a review of a selection of commonly studied boundary layer tripping methods. The main body of the thesis is devoted to the investigation of two novel passive flow control concepts, developed for this research, which were found to significantly reduce the drag coefficient in the sub-critical Re flow regime of a plain cylinder. Of these two concepts, the main research focus was on identifying the drag reducing mechanisms of a system of passive, continuously blowing jets. It was found that the interaction of the jets and cross-flow induced a very high frequency instability which leads to the downstream formation of tornado-like vortices which are shed into, and identified in the near-wake. It is postulated that the introduction of stream-wise vorticity into the separating shear layer develops favourable drag reducing mechanisms. Discrete cylindrical surface protrusions were additionally found to develop similar effects on flow topology and drag reduction, and were more effective than the passive jets at low Reynolds numbers.

Shock Wave Boundary Layer Interactions (SWBLIs) represent complex flow phenomena that remain poorly understood despite their prevalence on high-speed vehicles, in part due to their complicated underlying physics. In particular, the mechanisms that drive the high-amplitude, low-frequency unsteadiness within the interaction have perplexed researchers for many years while remaining a limitation to vehicle performance and a potential danger to airframe integrity. This investigation has specifically examined the influence of interaction strength and configuration type on the characteristic unsteady behaviour that describes the flow environment. Until now, researchers have typically focused on testing a specific configuration in a given test facility. This approach can obscure meaningful conclusions that may be drawn due to the interference of the test environment. The present research effort therefore tackles this flaw by assessing flow behaviours across a range of SWBLIs, all tested within a common environment. Four strengths of oblique shock reflection interactions and two strengths of compression ramp interactions have been assessed and compared. Experiments have been conducted in the Imperial College Supersonic Wind Tunnel with a Mach 2 turbulent incoming boundary layer with momentum thickness Reynolds number of 8000. Using a combined approach of synchronised PIV and fast-response wall-pressure measurements the unsteady elements to the interactions have been investigated. The spectral evolutions of unsteady wall-pressure disturbances are assessed throughout each of the interactions. Results confirm that the high-frequency component of the separation shock spectral content is common across all interactions. Meanwhile, low-frequency amplitudes scale with the interaction length, acting to decrease the characteristic frequency used to describe such motion when the interaction strength is increased. Instantaneous shock structures have also been identified which confirm the presence of two unsteady mechanisms governing the dynamics of the separation shock: rotation and translation. Quasi-steady modelling of these mechanisms indicates how their relative dominance varies with interaction strength and configuration type. This body of work represents a unique assessment of valuable data that is crucial to the development of unsteady SWBLI understanding.

This research examines the extent to which simulation can be used to train pilots of the Royal Air Force to fly and fight the Eurofighter Typhoon, and is the culmination of a series of trials over a period of 4 years. The approach was threefold, firstly examining the performance of students trained entirely on the Operational Conversion Unit’s full syllabus in the simulator and then tested against their peers on each of the four phases in live flight, secondly investigating the cultural acceptance levels of the present Typhoon pilots and lastly using lessons learnt to generate and test a syllabus to train Typhoon pilots to Mult Role Combat Ready in 40% of the present time. It was found that increasing the proportion of synthetics from the lowest Live Synthetic Balance (LSB) of 75:25 used on the front-line meets a cultural and resource barrier at 50:50. This did not represent the maximum LSB achievable however with the heavily synthetic Multi-Role Syllabus reaching an LSB of 21:79 with successful completion of the end of course test. Cultural acceptance of the simulator had correlations with the squadron a pilot was assigned to, the manner in which the simulators were programmed for use and the experience level of the pilot. No evidence was found within the sample to suggest age had an effect. Recommendations on minimum proportions of live and synthetic training was mapped for each of the required tasks and comparisons of these were made across complexity levels. Resource savings found by the generating and testing a Multi Role Combat Ready syllabus that recognised and incorporated all the strengths, weaknesses and lessons identified in the previous trials generated a saving of approximately 9 months and 100 Typhoon live flying hours per student, equivalent to approximately 1300 man maintenance hours that could be reinvested into personnel in the form of leave, adventurous training or development.

Attempts to predict the operating temperatures of disc brakes have shown the need for an investigation into the heat transfer from a rotating disc when the airflm" induced by rotation is disturbed in a number of ways, each linked with the conditions of the operating environment. A survey of previous work revealed that a considerable body of work existed on the basic systems of a disc rotating in still air and a rectangular flat plate in a uniform stream, which aided the understanding of the effects of more complex flow systems on the heat transfer, such as were studied durirthis investigation. Experiments were made with an electrically heated rotating disc; measurements were made of the surface temperature and heat input when a steady state was reached, and the heat transfer coefficient calculated from these. Experimental data were found for the heat transfer from a disc rotating in still air and in an airflow parallel and adjacent to the disc surface. The effects of masking certain sectors of the disc were found in both of the environments just described. Finally, experiments were made with boundary layer tripping devices and jets of air directed at the disc surfaces to discover the effectiveness of these methods in increasing the heat transfer. The results from the disc rotating in still air agreed with measurements by previous workers and a prediction based on a combination of data for a rotating disc and a stationary surface achieved good correlation with data recorded with the disc rotating in an air crossflow. The other sections of work, with masked sectors, boundary layer trips and air jets, introduced flow patterns which affected the heat transfer in a complex manner. These effects were explained with the aid of flow visualization and general theoretical correlations obtained.