This thesis addresses the problem of estimating rainfall rates from satellite imagery. The potential for using cloud motion and texture to estimate rain rates has been examined. The main types of textural information, i.e. statistical, structural, frequency and spatio-temporal, have been used to derive features from the satellite measurements and then used to determine a relationship to the radar-observed rain rates. These features were ranked by two scoring functions that were devised to assess their relationship to rain rates. The first scoring function selected a feature set for classifying samples into three rain rate classes. The selected features successfully classify rain rates of a mid-latitude cyclone seen on Meteosat7 with 84.8-99.3 % accuracy with a significant Hanssen-Kuipers discriminant score when a probabilistic neural network was used. A similar accuracy was found when a support vector machine (SVM) was used. Another scoring function was used for the selection of the features for estimating rain rates of each class. A Gaussian-kernel SVM that has been trained by these features produced visually agreeable rain estimates, which were much better than those produced by other methods that used only spectral information. Using the same types features at different time also achieved the similar accuracy. The method was robust and continuous rain estimates were obtained. Unlike other techniques in which additional information has always been required, the results showed that textural information alone can be used for rain estimation. This is preferable when only satellite measurements are available. Frequent updating of the observed rain rates can be done to improve the accuracy of the estimation. The potential for using cloud motion to predict rain rates was also examined. It was found that a combination of the maximum cross correlation and optical flow techniques provided the best estimate of the velocity of clouds. A cloud’s displacement derived by the maximum cross correlation technique was used for the approximation of the future location of its corresponding rain and the final velocity derived by the optical flow technique predicts how the rain rates would change. The rain rates predicted by this novel method provided good correlation to the observed rain rates at an hour later.

A wake behind an unsteady moving submerged vehicle is of interest and importance in a broad variety of engineering disciplines, ranging from underwater to aeronautical engineering. When the vehicle changes its speed or direction, under certain conditions, it can lead to the appearance of a coherent kilometre-scale quasi-planar counter-rotating vortical structure which persists for the order of days. The aims of this work are to determine the conditions under which such a large coherent vortex can appear and to obtain deeper understanding of its dynamics by investigating the evolution of a turbulent patch created by either an impulsively accelerating axisymmetric self propelled body or an impulsive jet in the small-scale upper ocean via direct numerical simulation. A non-conservative body force is applied to the governing equations to represent an impulsive jet, while an accelerating motion of a self-propelled body is emulated by the combination of an immersed boundary method and the body force. Criteria for the occurrence of a vortex dipole are found to depend on a dimensionless parameter, called the confinement number. Once the confinement number is higher than about unity, the vertical growth of an impulsively generated turbulent patch is restricted by the top and bottom layers of the upper ocean leading to the formation of a vortex dipole at the free surface. The contrast and strength of a surface signature increase linearly with increasing confinement number. The late-time dynamical structures, i.e. the propagation velocity, size and the decay rate of maximum vorticity, of the dipolar eddy induced by the presence of vertical confinement can be predicted by scaling laws relevant to a stratified fluid, even though the dipole possesses a Reynolds-number dependence.

No description available.

Evacuation modelling technology offers designers and regulators of aircraft new opportunities to rigorously test designs and theories. However, before evacuation models can be used effectively they need to be understood by the regulatory and aviation industry, validated and developed further. This thesis tackles each of these aspects. This thesis provides a detailed review of evacuation modelling with special emphasis on aviation evacuation models and the available data upon which models and understanding can be based. Of these the airEXODUS model is selected for this thesis and it is described in detail and critically evaluated. The evaluation revealed three main issues that needed to be addressed in order for aircraft evacuation modelling to advance. These issues relate to, (1) the limited quantity of model verification, (2) the inability of models to represent crew procedures, and (3) the limited behavioural capabilities of these models with regard to simulating real accidents as opposed to certification scenarios. The fundamental accuracy and predictive capability of airEXODUS is evaluated. This is followed by a comprehensive investigation of cabin crew and passenger behaviour in 90-second certification trials and real emergency evacuations. The conclusions from this investigation serve as the basis for the development of new algorithms to addresses issues (2) and (3). Behavioural algorithms are developed to simulate cabin crew bypass in conjunction with algorithms for passengers exit choice and methods for simulated passengers to optimise their chosen route to an exit. Finally, this thesis concludes by demonstrating the value of evacuation modelling in the design of future aircraft, the regulation of current aircraft and in understanding some of the contributing factors involved in past evacuation related disasters.

The level of ownership and use of motorcycling has increased rapidly in Edinburgh and the UK in the last ten years. In this study, motorcycle driving cycles (rural and urban) were developed for Edinburgh (Edinburgh Motorcycle Driving cycle-EMDC). The analysis of EMDC demonstrates that motorcycles‘ driving behaviour differs between urban and rural areas. EMDC shows a typical transient nature of speed, acceleration and deceleration, which is also different from regulatory driving cycles (Economic Commission for Europe-ECE and World Motorcycle Test Cycle-WMTC) and examples from Asia (Taiwan, Bangkok and China). This research underlines the need for detailed investigations of driving cycles in any local condition. It is not generally feasible for a driving cycle developed in one area to be applicable in another area, even with some similar characteristics. Emission factors were also estimated using onboard, laboratory and micro simulation measurements along the test corridor (Air Quality Management Area-AQMA). Laboratory measurements were carried out by applying a number of standard driving cycles (ECE and WMTC) and the derived EMDCs. Results show that the emission factors (EFs) calculated in the laboratory for carbon monoxide (CO) and Hydrocarbons (HC) are higher for the urban EMDC cycle compared to the standard regulatory factors than they are for the rural (except Nitrogen Oxide-NOx). Laboratory emission factors for CO and HC for the urban EMDC were found to be higher than the micro-simulation and onboard methods. EFs obtained from micro-simulation and onboard emissions using the National Atmospheric Emission Inventory (NAEI) emission coefficients were not very different with the exception of NOx, which were relatively higher than those of EMDC. Micro simulation models were mainly developed for private cars and therefore special care should be taken when using them for modelling other conditions (e.g. motorcycles driving characteristics). This study illustrates the extent to which micro-simulation may be utilised to accurately model emissions and discusses the refinements required to model motorcycle motion (hence emission) accurately in micro simulation. The study provides a platform for a large number of potential future applications for the evaluation of emissions and for developing various policy scenarios of pollution reduction and reducing health impacts at local levels.

Most drivers are aware of the distracting effects and possible dangers of using their mobile phones while driving. Many studies have investigated this effect and although not all have found large changes in driver performance, all have concluded that nothing good can come from this behaviour. Without adding to the wealth of studies in this area, this PhD aimed to discover what kinds of attitudes drivers have towards these technologies, how these drivers behave in a ‘natural’ situation and how these drivers can be persuaded to cease using these technologies while driving. The first study was a large questionnaire aimed at measuring theory of planned behaviour constructs from drivers towards five in-car technologies; hand-held mobile phone use, hands-free mobile phone use, text messaging, MP3 player use and satellite navigation use. These identified that attitudes in particular are important predictors of intentions and behaviour. Regression models explained 38-76% of the variance in past behaviour for the five technologies showing that the theory of planned behaviour is a useful tool in understanding these kinds of behaviour. Studies 2 and 2.5 aimed to discover whether in-car technology could have a positive effect on driver performance with the use of a ‘risk monitor’. These results indicated that heart rate could be influenced in a simulator by instilling a sense of risk. Although no differences in driving variables were found, participants’ heart rates increased as accident risk increased, importantly showing that feelings of risk arousal can be induced in a simulated environment. Studies 3 and 4 used a novel design in which participants were given the choice to interact with an MP3 player and skip music that they personally did not like. Study 3 found that participants spent a significant amount of time looking away from the road when interacting with the device. However, no differences in lane deviations or speed deviation were found. Study 4 was then conducted to increase sample size and increase variability by introducing a second speed limit. Results showed that participants had a tendency to miss this speed limit due to interactions with the device. Main effects were found for standard deviation of speed, with the easy skip condition showing more variability than the control condition, possibly reflecting compensatory behaviours due to technology use. Eye data also showed the effects of technology use where glance durations were significantly longer while skipping compared to controls and were significantly longer during hard compared to easy tasks. Study 5 aimed to combine the risk monitor studies with the MP3 studies and create an environment where participants were free to interact with the device if they wished, but were also given some information on their risk of accident to take into account. Drivers on low risk roads interacted with the device more often than on high risk roads and spent more time listening to unpleasant music when the risk tone was present. In terms of driving measures, participants reached higher speeds when the tone was not present, while mean speeds showed a tendency to be lower when the tone was present. Study 6 was an intervention study which, informed by the results of study 1, created an intervention based on changing drivers attitudes towards technology use while driving. After the intervention was given, drivers completed a similar study to the MP3 experiments. This meant that not only could attitudes be measured at different time points, but actual behaviour relating to technology use could be measured. Theory of planned behaviour interventions are often criticised as they tend to measure behaviour by questionnaire rather than tangible measures of behaviour and so this was an advantage to this study. However, the intervention showed no significant effect on attitudes or MP3 music skipping behaviour. However, a significant correlation was found between the number of times a participant skipped and measures of driving violations. The challenge for future work will be to determine how attitudes can be changed towards technology use and whether further experiments may observe ‘natural’ behaviour in relation to different types of technology.

This thesis involves the computation of aerodynamically generated sound using a source-extraction based coupling approach. In the present coupling method, the unsteady aerodynamic calculation and the calculation of sound propagation are separated artificially. A set of acoustic perturbation equations is derived by decomposing all flow variables into their dominant part and their fluctuating part, and neglecting some small-magnitude terms, and further simplified into a set of isentropic perturbation equations. Accompanying the derivation of the acoustic perturbation equations, a new extracting formulation for the acoustic source terms contained in the unsteady flow field is proposed. The acoustic source terms required in solving the acoustic perturbation equations are computed numerically from the time-dependent solutions of the unsteady flow field. In the simulation of the unsteady flow, the unsteady Reynolds-Averaged-Navier-Stokes equations (RANS) based cell-centred finite volume method is mainly used. A large eddy simulation (LES) technique is also employed in the investigation of one application case. A powerful and efficient high order dispersion-relation-preserving (DRP) finite difference scheme with fully staggered-grid variable arrangements is implemented in the solution of the acoustic perturbation equations. The performance of a set of radiation boundary conditions is examined for various background flows. A suitable and efficient coupling procedure, in conjunction with the source-extraction formulation, is designed between the cell-centred finite volume based CFD solver and the fully-staggered finite difference based acoustic solver. A range of acoustic model problems are investigated with the purpose of assessing the feasibility and accuracy of the source-extraction formulation associated with the coupling procedure. These model problems include wave propagation, reflection, interaction, and scattering, of acoustic pulse with/without background mean flow. The accuracy of computational results from these model problems is very encouraging when reasonable computational mesh sizes and time steps are used in both the CFD solver and the acoustic solver. Several applications of the source-extraction based coupling method to some more complex cases have also been examined. These cases are: 1) generation and propagation of sound by a series of vortices impinging on a finite thin flat plate; 2) generation and propagation of sound from a subsonic flow past a finite thin flat plate with a small angle of attack; 3) generation and near field radiation of aerodynamic sound from an low speed, laminar flow over a two-dimensional automobile door cavity; 4) flow-induced noise from an open cavity turbulent flow. These application calculations have demonstrated preliminarily the capability and potential of the new source extraction formulation for solving more realistic aeroacoustic problems.

This thesis investigates the application of artificial neural networks (NN) in air/fuel ratio (AFR) control of spark ignition(SI) engines. Three advanced neural network based control schemes are proposed: radial basis function(RBF) neural network based feedforward-feedback control scheme, RBF based model predictive control scheme, and diagonal recurrent neural network (DRNN) - based model predictive control scheme. The major objective of these control schemes is to maintain the air/fuel ratio at the stoichiometric value of 14.7 , under varying disturbance and system uncertainty. All the developed methods have been assessed using an engine simulation model built based on a widely used engine model benchmark, mean value engine model (MVEM). Satisfactory control performance in terms of effective regulation and robustness to disturbance and system component change have been achieved. In the feedforward-feedback control scheme, a neural network model is used to predict air mass flow from system measurements. Then, the injected fuel is estimated by an inverse NN controller. The simulation results have shown that much improved control performance has been achieved compared with conventional PID control in both transient and steady-state response. A nonlinear model predictive control is developed for AFR control in this re- . search using RBF model. A one-dimensional optimization method, the secant method is employed to obtain optimal control variable in the MPC scheme, so that the computation load and consequently the computation time is greatly reduced. This feature significantly enhances the applicability of the MPC to industrial systems with fast dynamics. Moreover, the RBF model is on-line adapted to model engine time-varying dynamics and parameter uncertainty. As such, the developed control scheme is more robust and this is approved in the evaluation. The MPC strategy is further developed with the RBF model replaced by a DRNN model. The DRNN has structure including a information-storing neurons and is therefore more appropriate for dynamics system modelling than the RBF, a static network. In this research, the dynamic back-propagation algorithm (DBP) is adopted to train the DRNN and is realized by automatic differentiation (AD) technique. This greatly reduces the computation load and time in the model training. The MPC using the DRNN model is found in the simulation evaluation having better control performance than the RBF -based model predictive control. The main contribution of this research lies in the following aspects. A neural network based feedforward-feedback control scheme is developed for AFR of SI engines, which is performed better than traditional look-up table with PI control method. This new method needs moderate computation and therefore has strong potential to be applied in production engines in automotive industry. Furthermore, two adaptive neural network models, a RBF model and a DRNN model, are developed for engine and incorporated into the MPC scheme. Such developed two MPC schemes are proved by simulations having advanced features of low computation load, better regulation performance in both transient and steady state, and stronger robustness to engine time-varying dynamics and parameter uncertainty. Finally, the developed schemes are considered to suit the limited hardware capacity of engine control and have feasibility and strong potential to be practically implemented in the production engines.

Tonal noise, the self-induced discrete frequency noise generated by aerofoils, is investigated. It is heard from an aerofoil placed in streams at low Mach number flows when inclined at a small angle to the stream. The tones are heard as a piercing whistle, commonly up to 30 dB above the background noise level. The work is motivated by the occurrence of tonal noise from rotors, fans and recently wind-turbines. Previous authors have attributed tonal noise to a feedback loop consisting of a coupling between laminar boundary-layer instability waves and sound waves propagating in the free stream. The frequency has been predicted by use of various methods based on this model. In this thesis a review of wind-tunnel results obtained by Dr. E.C. Nash at the University of Bristol is presented. Boundary-layer measurements show the presence of tonal noise is closely related to the existence of a region of separated flow close to the trailing edge of the aerofoil. Highly amplified boundary-layer instability waves were observed close to the trailing edge of the aerofoil at the frequency of the tone. A comprehensive analysis of the linear stability of the boundary-layer flow over the aerofoil is presented. The growth of boundary-layer instability waves over the aerofoil is calculated. The growth rates of the waves were obtained by solving the Orr? Sommerfeld problem at several stations on the aerofoil. The Falkner?Skan boundary layers were found to be a suitable form of velocity profiles to incorporate the adverse pressure gradients experienced by the flow over an aerofoil. The amplification of the instability waves is shown to be controlled almost entirely by the region of separated flow close to the trailing edge. The calculated frequency of the linear modes with maximum amplification over the aerofoil is found to be close to the observed frequency of the acoustic tone. A weakly nonlinear stability analysis was also performed and this appears to be a suitable description of the boundary-layer instability waves. The results indicate that the frequency of the tones may commonly be predicted to within 10% by using weakly nonlinear stability theory. The generation of sound by diffraction of the boundary-layer instability waves at the trailing edge of the aerofoil is also discussed as well as the proposed feedback models. A modified feedback model is proposed, being based on the experimental and theoretical results.

Each year millions of people around the world are killed or injured due to being involved in collisions while driving on the roads, with young and inexperienced drivers most likely to be killed or injured. Numerous studies have found a link between the likelihood of a driver being involved in a collision and their hazard perception (HP) ability, with young and inexperienced drivers having inferior HP abilities compared to older and more experienced drivers. This thesis presents a series of studies that investigate the factors that affect HP performance as well as comparing different HP testing methods. The traditional video-based method is explored as well as new methods utilising a high-fidelity driving simulator (used as an analogue of real-world driving) in order to see if there are ways of making HP testing more representative of detecting and responding to hazards while driving on the road. This thesis also explores the use of functional near-infrared spectroscopy (fNIR) as a portable and flexible means of measuring dorsolateral prefrontal cortex (DLPFC) activity during driving. Comparisons between video-based HP testing (similar to that used in the UK HP test) and simulator-based HP testing revealed significant differences in psychophysiological and behavioural responses, indicating that video-based methods may not be representative of HP while driving on the road. fNIR was found to primarily measure task workload and to be a reliable means of recording DLPFC activity across a range of different driving tasks. The fNIR results also demonstrated that DLPFC was similar when performing a simulator-based HP test and a hybrid simulator method using simulation replays, suggesting that introducing elements of driving simulation may help bridge the gap between current video-based HP testing methods and HP while driving on the road. These findings have important implications for the theoretical and practical aspects of HP testing and the use of fNIR in driving research.