No description available.

A study of an inverted single-element wing with vortex denerators in ground effect is presented. Counter-rotating and co-rotating rectangular-vane type vortex generators with different device heights are studied on the suction surface of the wing. The primary application of this study is a front wing of race cars. This study comprises three main elements: 1) experimental study, 2) computational study and 3) multi-fidelity surrogate modelling. An experimental testing is performed in a wind tunnel equipped with a moving belt rig, studying time-averaged and unsteady aerodynamic characteristics. The experimental work performed shows that a use of vortex generators, notably of the counter-rotating sub-boundary layer vortex generator type, can be effective at controlling flow separation, with a resultant improvement in downforce within relatively low drag penalty. The results also reveal fundamental vortex characteristics for flow separation control. A computational study is performed by three-dimensional Reynolds-averaged Navier-Stokes steady simulations with the Spalart-Allmaras turbulence model. The computations are validated against the experimental results so as to provide confidence, with the validation exhibiting close agreement with the experimental results. The computational results complement the experimental results by highlighting the flow physics of how vortex generators can help control flow separation on an inverted wing in ground effect and how critical vortex generator type and size are for its effectiveness. An application of a force-based vortex generator source term model is also examined. A multi-fidelity surrogate modelling approach working with co-kriging regression and statistical techniques is presented, unsing the experimental and computational data sets. This approach exhibits an efficient improvement of surrogate models, compared to conventional surrogate models. Surrogate models of sectional downforce of an inverted wing with counter-rotating sub-boundary layer vortex generators in ground effect are constructed as a response of the ride height and incidence.

Theoretical studies are described on the response of certain structures to acoustic waves. The most general problem treated is the response of a long cylindrical shell periodically stiffened both longitudinally and circumferentially when excited by obliquely incident plane sound waves. Numerical calculations are given for a structure having similar characteristics to an aircraft fuselage. Calculations of the shell vibration and internal sound field are given. A simplified mathematical model, related to the phasing of aircraft propellers, is used to show the effect on the internal sound field due to a phase difference between opposite running plane waves incident on the shell. A procedure is described for calculating the acoustic loading on a rectangular flat plate set in a rigid wall and vibrating in sinusoidal modes. Results are given of the acoustic radiation damping. Wide frequency band space correlation measurements of the pressure field close to a two-inch diameter cold air jet are described. The results are shown to be consistent with the hypothesis that the pressure field close to the jet is convected at a speed related to the local speed in the jet. Using the convection hypothesis a generalised non-dimensional form of the space correlation is derived and both model and full scale measurements are shown to conform. The convection model predicts the form of the narrow frequency band space-correlation function. Tliis function is required for the calculation of structural response to jet noise. An approximate form of this function is utilised in a calculation of the response of a flat panel close to an air jet. Space correlation measureinents in the region within five jet diameters of the orifice of a full scale jet engine exhaust show the acoustic nature of the radiated noise showing the pressure field propagating at the speed of sound as opposed to the near pressure field of the jet turbulence propagating downstream at about one half of the local jet speed.

No description available.

This thesis introduces a novel framework for life cycle mission modelling during conceptual aeronautical design. The framework supports object-oriented mission definition using Geographical Information System technology. Design concepts are defined generically, enabling simulation of most aeronautical vessels and many non-aeronautical vehicles. Moreover, the framework enables modelling of entire vessel fleets, business competitors and dynamic opera-tional changes throughout a vessel life cycle. Vessels consist of components deteriorating over time. Vessels carry payload that operates within the vessel environment. An agent-based simulation model implements most framework features. It is the first use of an agent-based simulation utilising a Geographical Information System during conceptual aero-nautical design. Two case studies for unmanned aircraft design apply the simulation. The first case study explores how the simulation supports conceptual design phase decisions. It simulates four different unmanned aircraft concepts in a search-and-rescue scenario including lifeboats. The goal is to learn which design best improves life cycle search performance. It is shown how operational and geographical impacts influence design decision making by generating novel performance information. The second case study studies the simulation optimisation capability: an existing aircraft design is modified manually based on simulation outputs. First, increasing the fuel tank capacity has a negative effect on life cycle performance due to mission constraints. Therefore, mission definition becomes an optimisation parameter. Changing mission flight speeds during specific segments leads to an overall improved design.

Almost all automotive fuel injection systems are experiencing some form of cavitation within their nozzle under different operating conditions. In-nozzle cavitation initiates in various forms and directly influences the emerging spray. Experimental studies have shown that cavitation in diesel injectors leads to smaller droplet formation, especially by the on-going trend towards higher injection pressures, which enhances fuel evaporation but also creates undesirable consequences due to transient nature of cavitation such as spray instabilities, erosion on internal surfaces, and hydraulic flip. Thus, the understanding of the internal flow of automotive fuel injectors is critical for injector design. On the other hand, biodiesel has emerged as one of the potential alternative fuel which can also be carbon neutral because it uptakes CO2 during cultivation of its feedstock and can be used in existing diesel engines with little or no modifications. Therefore, the present study is focused on assessing and outlining cost-effective methods to analyse internal flow in fuel injectors for diesel and biodiesel fuel applications. In the present study, RANS-based (Reynolds-averaged Navier–Stokes) CFD (Computational fluid dynamics) approach has been chosen to simulate quasi-steady flows in the steady state test rigs of fuel injectors of IC engines. The RANS approach is selected over computationally expensive SAS (Scale Adaptive Simulations), DES (Detached Eddy Simulations) and LES (Large Eddy Simulations) because it was considered that these quasi-steady simulations could be performed within hours and with less computing resources using RANS rather using SAS, DES and LES which may require orders more time and computing resources. Cavitation models and RANSbased turbulence models have been evaluated for single-hole and multi-hole injectors operating on steady state test rigs. Furthermore, influences of liquid and vapour compressibility were also investigated. Influences of biodiesel properties such as higher viscosity and density on cavitation were also assessed. In the first part of the study, single-phase simulations have been carried out in the mini-sac type multi-hole (6) injector. Several two-equation turbulence and near wall models were assessed, amongst most appropriate for the application were identified. Predicted mean velocity and RMS velocity were compared with measurements and showed good agreements. Flow field analysis showed predictions of different types of vortices in the injector. Two main types of vortex structures were predicted: ‘Hole-to-hole’ connecting vortex and double ‘counterrotating’ vortices emerging from the needle wall and entering the injector hole facing it. The latter create a complex 3D flow inside the injector hole when it interacts with the recirculation region at the entrance of the injector hole. Cavitation simulations inside a single-hole injector were next performed. Simulations were assessed by comparing predicted vapour volume fraction with measurements. Influences of liquid and vapour compressibility were also checked. The compressibility of vapour was modelled using ideal gas law and liquid compressibility was modelled using the Tait equation. Vapour compressibility resulted in an increase of vapour volume fraction at the low-pressure region and predictions were also in better agreements with experimental data. The liquid compressibility made no impact on the simulation results. The local sonic speed in the liquid-vapour mixture was computed using Wallis model which predicted a very low local sonic speed in the liquid-vapour mixture. Therefore, the local flow in liquid-vapour mixture became supersonic. A normal shock wave was predicted just downstream of the cavitation bubble cloud as local flow velocity was reduced from supersonic to subsonic. Finally, the cavitation simulations were performed in the enlarged mini-sac type multi-hole injector. Established turbulence, cavitation and compressibility models from above studies were used. Reasonable quantitative agreements with experimental data were obtained for the mean axial velocity and RMS velocity. Reasonable qualitative agreements were also achieved when predicted cavitation results were compared with high-speed digital images. Henceforth a parametric study to assess the influence of biodiesel fuel properties such as an increase in viscosity and density on the cavitation was performed. Viscosity and density of both phases in the fluid were parametrically increased by 20%. Results showed that cavitation was suppressed when the viscosity was increased because it increased the flow resistance, thus reduced the velocity. This caused a reduction in the size of recirculation region at the entrance of the injector hole and hence a smaller saturation pressure region was predicted. Cavitation was further suppressed when density was increased causing the reduction in the velocity at the same mass flow rate, which further reduced the recirculation region, therefore, reduced the saturation pressure region and consequently cavitation.

The work presented in this thesis addresses issues involving the accurate and efficient numerical modelling of turbulence combustion with an emphasis on an industrially representative Tay model combustor. This combustor retained all essential features of a modern aero-engine rich burn combustor and thus the turbulence combustion within this combustor is much more complicated than those observed in the combustor-like burners typically considered in laboratory experiments. A comparative study of two combustion models based on a non-premixed assumption or a partially premixed assumption using the previously proposed models Zimont Turbulent Flame Speed Closure (ZTFSC) and Extended Coherent Flamelet Method (ECFM)) is presented in a first step. Comprehensive chemical reactions containing 244 reactions and 50 species are taken into account using a tabulated detailed chemistry approach and an assumed shape PDF to account for turbulence effects. The purpose of this study is to validate and compare the effectiveness of these models in predicting complex combustion and to improve upon for the defects observed in previous predictions of the same combustor. It is concluded that the use of models invoking the partially premixed combustion assumption can provide much more accurate results than models using a non-premixed combustion assumption especially in the primary zone of the combustor where turbulence combustion interaction is strong. In addition, certain shortcomings of steady RANS type models are identified as a result of strong unsteady effects and their inability to resolve the turbulence spectrum. Following this, two URANS models and the scale resolving simulation (SRS) approach such as a shear stress transport, K-omega, scale adaptive simulation (SSTKWSAS) combined with the partially premixed method identified in the first step are employed in a second step to further improve the accuracy achieved and to provide evidence and guidance in terms of the trade-off between accuracy and computational cost for complex turbulent combustion simulations. The second generation SRS model (SSTKWSAS) is applied to the complicated flow environment of a realistic combustor for the first time. The present work highlights the superiority of the combination of the SSTKWSAS approach and a partially premixed combustion model in terms of both accuracy and efficiency for predicting such combustion problems.

Strong ionospheric electron content gradients may lead to fast and unpredictable fluctuations in the phase and amplitude of the signals from Global Navigation Satellite Systems (GNSS). This phenomenon, known as scintillation, can impair the tracking performance of a GNSS receiver, leading to increased phase and Doppler errors, cycle slips and sometimes to complete losses of signal lock. In order to mitigate scintillation effects at receiver level, the robustness of the carrier tracking loop, the receiver’s weakest link under scintillation, must be enhanced. Thanks to their adaptive nature, Kalman Filter (KF) based tracking algorithms are particularly suitable to cope with the variable working conditions imposed by scintillation. However, the effectiveness of this tracking approach strongly depends on the accuracy of the assumed dynamic model, which can quickly become inaccurate under randomly variable scenarios. This research work shows how inaccurate dynamic models can lead to a KF suboptimum solution or divergence when both strong phase and amplitude scintillation are present. Then, to overcome this issue, two novel self-tuning KF based carrier tracking algorithms are proposed. They self-tune their dynamic models by exploiting the knowledge about scintillation, which is achieved by estimating a number of scintillation indices. These types of tracking schemes are particularly suitable for ionospheric scintillation monitor receivers, which are designed for the computation of scintillation indices and other related parameters. Moreover, this thesis analyses and implements algorithms for a reliable computation of scintillation indices even when low cost receivers are exploited. Furthermore, a technique is proposed to compute scintillation indices even if temporary losses of signal lock or cycle slips occur. All algorithms have been assessed by exploiting both simulated and real data affected by high latitude and equatorial scintillation. Results show that the proposed algorithms are able to maintain the signal lock and provide reliable scintillation indices when classical architectures and commercial Ionospheric Scintillation Monitoring Receivers (ISMRs) fail.

In this thesis, the problem of A/F ratio control in a natural gas, internal combustion engine is addressed, with the global objective of reducing exhaust emission pollutants. A review of some mechanical approaches to exhaust pollutant reduction are assessed. It is found that many techniques aid the reduction of exhaust pollutants, but the most effective is the 3-way catalytic converter. To maintain conversion efficiency, the A/F ratio must be strictly controlled within the catalyst window limits around the stoichiometric operating point. In order to investigate possible control techniques, a mathematical model is developed to simulate the physical behaviour of the engine processes. This approach allows a quick turn-around in terms of cost and time, for control investigations. The model demonstrates close trend-wise approximation of the engine states with previous modelling studies, however, a full validation study was not possible. The model is then used to conduct investigations into A/F ratio control through the process of simulation. Conventional Pl-closed-loop control is assessed for steady-state and transient engine conditions, and for varying microprocessor sampling rates. It is found that Pi-control effectively removes state estimation errors, but is unable to remove A/F ratio excursions under transient operation. An open-loop compensation control structure is then developed as an extension to the IM-controller action. Simulation results show this approach to drastically reduce A/F ratio excursions for a number of typical driving scenarios. Potential problems that could well be encountered in the “real” engine environment are then investigated, and the practicality of the new controller assessed. A new approach to control is simulated that affords the most appropriate state estimation for the modelled system. This is shown to improve A/F ratio control upon that of the conventional approach but cannot match the compensation controller ability.

Cost is one of the problems for wider adoption of Advanced Driver Assistance Systems (ADAS) in China. The objective of this research project is to develop a low-cost ADAS by the shared use of motion vectors (MVs) from a H.264/AVC video encoder that was originally designed for video recording only. There were few studies on the use of MVs from video encoders on a moving platform for moving object detection. The main contribution of this research is the novel algorithm proposed to address the problems of moving object detection when MVs from a H.264/AVC encoder are used. It is suitable for mass-produced in-vehicle devices as it combines with MV based moving object detection in order to reduce the cost and complexity of the system, and provides the recording function by default without extra cost. The estimated cost of the proposed system is 50% lower than that making use of the optical flow approach. To reduce the area of region of interest and to account for the real-time computation requirement, a new block based region growth algorithm is used for the road region detection. To account for the small amplitude and limited precision of H.264/AVC MVs on relatively slow moving objects, the detection task separates the region of interest into relatively fast and relatively slow speed regions by examining the amplitude of MVs, the position of focus of expansion and the result of road region detection. Relatively slow moving objects are detected and tracked by the use of generic horizontal and vertical contours of rear-view vehicles. This method has addressed the problem of H.264/AVC encoders that possess limited precision and erroneous motion vectors for relatively slow moving objects and regions near the focus of expansion. Relatively fast moving objects are detected by a two-stage approach. It includes a Hypothesis Generation (HG) and a Hypothesis Verification (HV) stage. This approach addresses the problem that the H.264/AVC MVs are generated for coding efficiency rather than for minimising motion error of objects. The HG stage will report a potential moving object based on clustering the planar parallax residuals satisfying the constraints set out in the algorithm. The HV will verify the existence of the moving object based on the temporal consistency of its displacement in successive frames. The test results show that the vehicle detection rate higher than 90% which is on a par to methods proposed by other authors, and the computation cost is low enough to achieve the real-time performance requirement. An invention patent, one international journal paper and two international conference papers have been either published or accepted, showing the originality of the work in this project. One international journal paper is also under preparation.