Aircraft nose landing gear must be tested for shimmy oscillations experimentally. Current test methods consider the landing gear as stand alone systems without the possibility of accounting for the dynamics of the fuselage which they are attached to. This means that all possible interaction between them is ignored during the experiment which, if the interaction for specific conditions is significant, might reduce the reliability of the test results. On the other hand, conducting a full scale experiment on the aircraft would be impractical due to size and related costs. In this thesis the nonlinear interaction between the nose landing gear and the airframe is explored and the feasibility of the real-time dynamic substructuring test method to this particular system is discussed. Due to the nonlinearities involved, the applied method in the majority of this work is bifurcation analysis. First, a coupled nose landing gear-fuselage model is developed. It is a simplified two degree-of-freedom nose landing gear model coupled to the lateral component of the fuselage motion at the attachment point. By means of one- and two-parameter bifurcation diagrams it is demonstrated how the presence of fuselage dynamics may influence the behaviour of the landing gear - and vice versa, how the vibrations of the gear affect the dynamics of the fuselage - when key parameters of the system are varied. It is found that dynamic interaction between the two subsystems is possible. How significant this interaction is strongly depends on the characteristics of the fuselage dynamics. Owing to this interaction the application of the real-time dynamic substructuring method on the model is reasonable and, hence, its feasibility is explored numerically and analytically. In a substructuring test part of the system is kept as the physical test specimen while the remainder is simulated numerically. The numerical and physical substructures are connected via a transfer system (an actuator) and the interface is controlled in real-time to ensure that the dynamic behaviour of the substructured system replicates with high accuracy that of the system being emulated. The main challenge in a substructuring test is the delay arising from the response time of the actuator. The effect of the delay is analysed in terms of the proposed test configuration, which consists of the fuselage as the numerical model and the nose landing gear as the physical substructure; the latter is also represented by a simplified two degree-of-freedom model. It is found that the substructured system is extremely sensitive to the delay. In order to reduce the sensitivity, without modifying structural parameters, the numerical model is extended by an additional compliance, and the influence of this on the sensitivity is studied. By careful adjustment of the additional compliance, the robustness of the test can be improved; however, the accuracy may then be compromised.

A wide range of Unmanned Air Vehicle applications have been identified over the last decade, both for civilian and military usage. The Air-to-Air Refuelling capability is perceived as an advantage to future support and deployment of mid-large scale Unmanned Air Vehicles to meet operational requirements. Despite Autonomous Air-to-Air Refuelling drawing a lot of attention and research in general, bow wave effects in an Air-to-Air Refuelling scenario have received little attention. Some existing studies have discussed and attempted to model the effect of the bow wave from a comparatively large receiver (B-2, C-17) onto the tanker (KC-135) during boom refuelling. However the effects of bow wave from the receiver aircraft onto the drogue in a probe and drogue refuelling system have received very little attention. This thesis analyses the effects of the receiver's bow wave on the drogue in a probe and drogue based Air-to-Air Refuelling activity, with results suggesting its critical influence on the capture rates of Autonomous Airto- Air Refuelling simulation. This work has developed a state of the art literature survey of Air-to-Air Refuelling including the developments in the sensor models, numerical modelling concepts, control methodologies and simulation and test facilities. By performing multiple random simulations a novel way of quantifying capture rates in aerial refuelling simulations has been developed. Compensating the bow wave effects through the use of position offset is simple yet effective solution emerging from this work. This method was found to restore the loss in performance in the Autonomous Air-to-Air Refuelling simulation which results from the bow wave interactions. On the other hand, the ability to optimise these position offsets for a given condition makes it suitably applicable to different flight conditions. Another major contribution is the various capture strategies presented in this thesis which demonstrate several ways of approaching the drogue in the capture phase for successful engagements. Results demonstrate further gains in terms of capture rates by avoiding drogue chasing in the simulations. Additional unconventional and innovative concepts to drogue capture are also discussed. This work forms part of the Autonomous Systems Technology Related Airborne Evaluation & Assessment programme in the UK.

Bird strikes represent a major hazard to the aerospace composite structures, due to their low impact resistance. Accurate selection and lay up of the materials in the composite structure can significantly improve the out of plane properties of the composites. However, application of the complex hybrid sandwich composites into bird strike proof structures was not investigated yet. Therefore, this work was focused on the soft body impact resistance of a novel composite design for aerospace applications. The investigation was divided into experimental and modelling parts. In the beginning of this thesis, the numerical techniques for modelling of bird im¬pact and composite materials were studied. The theoretical background for the corresponding issue was provided, followed by the thorough validation of the exist¬ing numerical approaches. A Smooth Particle Hydrodynamic (SPH) method was chosen for the modelling of the soft body. This modelling technique was validated against experimental data for the rotating fan blade. Three parametric studies of bird impacting fan blades revealed strong influence of the bird impact location and timing on the final deformed shape of the blade. Moreover, it was proved that the SPH is capable of reproducing the exact load on the structure and is appropriate technique for modelling bird strikes.

There are many analytical models to predict turbofan noise radiation in a free field. However, these models explicitly ignore the effect of the aeroplane components on the engine noise. The effects of the aeroplane on the radiated noise must be included because it is the installed engine that affects people in the cabin and community. The original contribution of this thesis is to present a theoretical model for the near-field fuselage installation effects on tonal noise radiating from a turbofan engine inlet. Historically, the fuselage installation effects have been modelled using theoretical methods for open-rotor type sources. Installation effects include the scattering effect from the fuselage, and the refraction effect of the boundary layer running down the fuselage. In this thesis the established techniques are extended to include a new sophisticated analytical source for spinning modes radiating from a circular duct. The source model includes the diffraction effect of the duct lip. The model applies Fourier methods and implements the Wiener-Hopf method for diffraction. Owing to the physics of the problem, simplications in geometry and flow do not curtail the validity of the predictions. The refraction effect was quantified by evaluating the difference between sound pressure levels with and without the boundary layer. Upstream of the source the refraction leads to a paucity of surface acoustic pressure, yet downstream the boundary-layer effect was minimal. Further investigations led to an alternative method of simulating refraction by altering wavenumbers in Fourier space. The installation method was optimised for a turbulent boundary-layer profile by replacing a power-law with a scaled step-change boundary-layer profile. The model developed in this thesis combines the most sophisticated analytic radiation models with current installation models. Due to the speed of the method, the intended purpose for industry is to refine variables via parametric studies. Once these are established, a numerical method could include more complex geometry and flow to the model. By calculating the noise on the outside of the fuselage, the quantity and distribution of acoustic lagging can be optimised. This, over the lifetime of an aeroplane, could lead to appropriate noise levels in the cabin whilst achieving potential reductions in fuel consumption, emissions and costs.

Appraisal of the noise produced at the combustion stage in a jet engine is becoming more important, as fan and jet noise have been significantly reduced over many years. Therefore, combustion noise is contributing more to overall noise, especially at low jet velocities. Environmental regulations stipulate that gas emissions from a jet engine should be reduced. Thus, new techniques have been introduced in their operation, especially concerning the combustion process. Accordingly, there is a need for improved processing methods in order to extract combustion noise from other sources in new build engines. A novel processing technique to extract turbofan engine combustion noise called 3S-Array is presented. It has been developed using a multiple coherence technique with data acquired in the in-duct and external sound fields of a jet engine. In-duct sensors are located in the combustion chamber and in the nozzle of the engine, and external data is acquired using an array of microphones. A beamformed signal focused on the nozzle of the engine is generated with the data from the external array. Jet noise and in influences of the room on the array output are reduced using this focusing technique, which is referred to as Focused Beamformed Output (FBO). Results show that using this new 3S-Array technique with two of the in-duct sensors and the focused beamformed signal as the third one, provides a better estimator of combustion noise than the 3-signal coherence technique alone, or the Coherence-Output Power Spectra (COP), both of which are reported in the literature as methods for the extraction of combustion noise from the radiated noise spectrum.

The provision of electrical power on modern aircraft is a necessary and growing aspect of a gas turbine's function. The replacement of traditional pneumatic, hydraulic and mechanical systems with electrical equivalents means that electricity is now the dominant means of power distribution on aircraft. However, the electrical loads seen on aircraft present challenges, as they are time varying and are often non-linear. This is particularly true for loads such as radar. The aviation industry has adopted the term More Electric Aircraft (MEA) to describe the latest generation of aircraft with a high reliance on electrical power. There is potential for significant interaction between the transient variation of electrical loading and the gas turbine (both drive-train and engine core). Engine testing and initial simulation work support this view. Understanding of this phenomenon must now be furthered through modelling and testing. This thesis presents simulation models of a transmission system and generator interface, which provides a useful kernel for a modelled system to assess electro-mechanical interaction. This is extended to multi-domain simulation work through the successful interlinking of transmission, generator and an electrical load model. These models have been validated, at a domain level, against analytical expressions, and also as a complete electro-mechanical system against test data. To allow more control over test conditions, an electro-mechanical test rig is designed and constructed. The data from the test rig is analysed and compared to modelled results. This thesis also presents potential mitigation actions for avoiding unwanted electro-mechanical interactions during electrical load transients. A method of extracting transient mechanical torque information from a gas turbine's electrical generator's terminal quantities is included. At a system level, the simulation work in this thesis potentially enables the development of future designs with improved power systems integration throughout the entire airframe. High level control could allow optimisation of the power conversion process between gas turbine spool and electrical systems, with increased intelligence in the movement of power between components.

This thesis combines experimental and computational methods to investigate the low Reynolds number flow (Re = 50,000) around a NACA 4415 aerofoil, and its control using periodic surface motion. A physical model was fabricated and tested in a closed-loop wind tunnel and a good comparison between the experiments and computations was achieved. Time-resolved measurements of the surface reveal that the peak-to-peak displacement is a function of both the amplitude and frequency of the input voltage signal but the addition of aerodynamic forces does not cause significant changes in surface behaviour. The vibration mode shape exhibits a single peak and is uniform in the spanwise direction at frequencies below 80Hz, above which a change in the vibration mode occurs. The flow around the actuated aerofoil was compared with the baseline (i.e. unactuated) flow. The latter exhibits a large separation region and, as a result, produces relatively high drag and low lift forces. By analysing the experimental and computational data, the large separation zone was found to be the result of laminar separation without reattachment. Transition to turbulence does occur but too close to the trailing edge, and far from the wall, for sufficient pressure recovery to take place for reattachment. When actuated at 70 Hz, the frequency spectra in the vicinity of the trailing edge and near-wake was found to be dominated by the actuation frequency. Sharp peaks suggest the production of Large Coherent Structures at this frequency. In agreement with the experiments, the computations revealed that the vortex shedding from the shear layer was 'locked-on' to the surface motion and spanwise coherent vortices were produced during each actuation cycle. The increased momentum entrainment associated with them enabled a large suppression of the separated region, which was seen in both the experiments and computations. The result was a simultaneous increase in Lift and decrease in Drag and therefore a large increase in the L/D ratio.

The growth in competitiveness in airline industry has called for more advanced tool to estimate the operating costs. Engine maintenance costs are an important decisionmaking element during airline fleet selection judgment. Long term observation in aerospace led to the development of engine maintenance costs calculators based on empirical correlation. But the possibilities of empirical model application for future engines without prior operational data are limited. A physics-based tool to estimate the life of the engine components and predict the shop visit rate requires the variations of thermodynamic parameters over the flight path. High fidelity engine models are simulated using an engine performance program. A test program designated for design, off-design and transient performance simulation for simple turbojet layout gas turbine engine has been programmed and tested. The knowledge gained from program coding was used to generate more robust transient performance code implemented to Turbomatch. Two transient methods have been tested: The rapid transient performance method and the thermodynamic matching method. The tests showed greater robustness and stability of the second method, which has been finally adopted for the program. For industrial engine configuration and for future novel engine cycles the heat-exchanger dynamic response model was implemented and tested. Created tool was demonstrated on short-haul study of engine flight path analysis. Together with the aircraft model, the tool produced variations of parameters needed for the lifing algorithm.

This work focuses on the current civil engine design practice of increasing overall pressure ratio, turbine entry temperature and bypass ratio, and on the technologies required in order to sustain it. In this context, this thesis contributes towards clarifying the following gray aspects of future civil engine development: the connection between an aircraft application, the engine thermodynamic cycle and the advanced technologies of variable area fan nozzle and fan drive gearbox. the connection between the engine thermodynamic cycle and the fuel consumption penalties of extracting bleed or power in order to satisfy the aircraft needs. the scaling of propeller maps in order to enable extensive open-rotor studies similar to the ones carried out for turbofan engines. The rst two objectives are tackled by implementing a preliminary design framework, which comprises models that calculate the engine uninstalled performance, dimensions, weight, drag and installed performance. The framework produces designs that are in good agreement with current and near future civil engines. The need for a variable area fan nozzle is related to the fan surge margin at take-o , while the transition to a geared architecture is identi ed by tracking the variation of the low pressure turbine number of stages. The results show that the above enabling technologies will be prioritised for long range engines, due to their higher overall pressure ratio, higher bypass ratio and lower speci c thrust. The analysis also shows that future lower speci c thrust engines will su er from higher secondary power extraction penalties. A propeller modelling and optimisation method is created in order to accomplish the open-rotor aspect of this work. The propeller model follows the lifting-line approach and is found to perform well against experimental data available for the SR3 prop-fan. The model is used in order to predict the performance of propellers with the same distribution of airfoils and sweep, but with di erent design point power coe cient and advance ratio. The results demonstrate that all the investigated propellers can be modelled by a common map, which separately determines the ideal and viscous losses.

The principal aim of this research was to provide a detailed understanding of the performance of gas turbine engines inside indoor sea-level test beds. In particular the evaluation of both thrust correction factors and the estimation of the mass flow entering the test cell were at the core of the research. The project has been fully sponsored by Rolls-Royce pIc. Initially, their principal objective was to assess the relevance and accuracy of CFD when applied to thrust measurement inside indoor test beds with an intended outcome of minimising the use of expensive experimental measurements. The different system interfaces and accounting systems for in-flight conditions, available in the open literature have been developed and adapted for indoor environments. This has led to the definition of three different thrust correction equations using alternative definitions of thrust correction factor. Aero-dynamic principles have been applied for the derivation of one-dimensional relationships for the calculation of each thrust correction factor using generic engine-cell performance and dimensions. A one-dimensional analytical model has been developed to represent the enginedetuner ejector pump. This is able to characterise the engine-cell system performance and is used as the main tool for providing a matching procedure capable of predicting the cell entrainment ratio. By processing experimental data relevant to different engine-cell configurations through the ejector pump analytical model, a method for achieving the entrainment ratio control inside the cell has been identified. The CFD work has been concentrated into three main activities: • A quantitative extrapolation of the thrust correction factors including, the pre-entry force, the external and the total bellmouth force, the throat stream force, the intake momentum drag and the base drag. • The representation of the engine-detuner ejector performance for a variety of engine-cell configurations. • The modelling of the generic test cell components including the inlet stack, the cascade elbow, the exhaust stack & the blast basket. The outcomes of this research have been very successful in enhancing the validity of the thrust correction equations developed. In particular, the use of a one-dimensional approach in their estimation has been shown to be fully justified. The work has also emphasised the value of CFD in supporting the derivation of the matching procedure for predicting and controlling cell entrainment ratio. Indeed, one of the strongest outcomes of this work has been the conclusion that both the engine-cell characteristic lines computed with the one-dimensional model and those computed with CFD for different cell configurations are almost identical. In addition, the use of CFD as a tool for the quantitative evaluation of the thrust correction factors has been established. Finally, the CFD results have facilitated an enhanced understanding of the complex flow structure inside indoor test cells