No description available.

The automotive industry is currently driven to reducing fuel consumption in internal combustion (IC) engines and hence much research is being done into alternative fuels and power sources. Thermal energy recovery from IC engines has proved to be of considerable interest within the automotive industry. The motivation is that fuel consumption can be reduced with a minimal effect on the “host” technology of the vehicle. This thesis reports on a project that aimed to investigate the architecture and control of a thermal energy recovery system, working towards proving this novel system concept. This was achieved by the use of software modelling techniques and experimental tests on various components of the system, namely heat exchangers and steam expanders. Various modelling toolboxes were used to model a) a hybrid vehicle configuration and b) steam expanders. The hybrid vehicle modelling began as a basic model to demonstrate the hybrid application and configuration of the steam system, and was further developed to control and optimise the system in such a way that the fuel economy, the overall efficiency of the IC engine and the heat recovery system were all maximised. Standard drive cycles were used to run the hybrid vehicle models. The steam expander modelling was performed in order to validate the results from a series of experimental tests and also to deduce if the expander models could be scaled up to predict results for larger expanders. The fuel consumption for the initial modelling showed a reduction of between 8% and 36%, depending on drive cycle and modelling toolbox used. With the development of a simple PID controlled system, the fuel consumption was further reduced resulting in a range of 26% to 41%, again depending on drive cycle and modelling tool box. The experiments on steam expanders point to a uni-flow configuration being the most suitable. The expander modelling presents the groundwork for developing expander models to be used for validating the experimental results; again the uni-flow arrangement gave the most promising results. This thesis presents the results and draws conclusions from each project step; these conclusions are summarised together with some recommendations for future work.

Automotive manufacturers are investing in research and development for hybridization and more modern advanced combustion strategies. These new powertrain systems can offer the higher efficiency required to meet future emission legislation, but come at the cost of significantly increased complexity. The addition of new systems to modernise an engine increases the degrees of freedom of the control problem and the number of control variables. Advanced combustion strategies also display interlinked behaviour between control variables. This type of behaviour requires a more orchestrated multi-input multi-output control approach. Model based control is a common solution, but accurate control models can be difficult to achieve and calibrate due to the nonlinear dynamics of the engines. The modelling problem becomes worse when some advanced combustion systems display nonlinear dynamics that can change with time. Any fixed model control system would suffer from increasing model/system mismatch. Direct feedback would help reduce a degree or error from model/system mismatch, but feedback methods are often limited by cost and are generally indirect and slow response. This research addresses these problems with the development of a mobile ionisation sensor and an online adaptive control architecture for multi-input multi-output engine control. The mobile ionisation system offers a cheap, fast response, direct in-cylinder feedback for combustion control. Feedback from 30 averaged cycles can be related to combustion timing with variance as small as 0.275 crank angle degrees. The control architecture combines neural networks and fuzzy logic for the control and reduced modelling effort for complex nonlinear systems. The combined control architecture allows continuous online control adaption for calibration against model/plant mismatch and time varying dynamics. In simulation, set point tracking could be maintained for combustion timing to 4 CAD and AFR to 4, for significant dynamics shifts in plant dynamics during a transient drive cycle.

The problem of the effect of structural modifications on sound radiation from a structure is very important with many applications in the design of a variety of products. In practise, for the design of acoustically optimum structures simplified theoretical models are used which can lead to unexpected behaviour of the real structures. Alternatively, measurements or numerical simulation can be used. The disadvantage of these two methods is that the study of alternative structures (structures with different properties) is not easy since for every modification a new structure must be manufactured or a new numerical model must be created. Moreover, these two methods do not easily give an insight into the physical mechanisms of sound radiation from modified structures. In the first part of this thesis the problem of vibration and sound radiation from a plate with an attached beam stiffener is studied theoretically. This extends the current theories of vibroacoustic behaviour of infinite plates with an infinitely long beam discontinuity and infinite plates with an infinite number of equidistant beam discontinuities forming a periodic structure. Firstly, the propagation of flexural waves and the subsequent sound radiation from an infinitely long plate strip is considered. The scattering of plate flexural waves by a finite beam across its width is considered. Changes to the mean square velocity, sound power and radiation efficiency of the plate strip due to the introduction of the beam stiffener are identified. Simplified approximate analytical expressions for the low-frequency range, well below the critical frequency, are also presented. This model is extended for the case of a finite rectangular plate by incorporating wave reflection from the two additional boundaries of the plate. Expressions for the radiation efficiency and the mean squared velocity of the stiffened plate are derived. The results are compared with results derived using well established numerical methods. For structures with more complicated modifications the derivation of analytical expressions is not easy. For this reason numerical optimisation is often used. In this thesis numerical optimisation is used to optimise the modes of a structure in order to radiate acoustic energy weakly into the acoustic medium. The effectiveness of point mass and line stiffener modification to create acoustically optimum modeshapes in a flat simply supported plate is firstly studied. The results show significant reduction in the sound power radiated by the optimised structural modes. These results are also verified experimentally. This optimisation method has certain advantages when it is used on automotive panels. The main advantage is that the panel under consideration can be isolated from the rest of the automotive structure and it can be optimised alone. This drastically reduces the time required and makes the optimisation practically applicable. A simplified car model is used and the proposed optimisation is applied to one of its floor panels. Firstly, point masses and line stiffeners are used as structural modifications to create weakly radiating modeshapes. Then more commonly used geometrical modifications are used on the floor panel. Two such modifications are studied; swages and domes. The results show a significant reduction in the radiation effciency and sound power radiated by the optimised panel.

Envirom-nental pollution is a subject that needs urgent addressing. Since the internal combustion engine has its fair share of accountability on this, research on techniques for increasing engine efficiency and emissions is necessary. Controlled Auto Ignition is a promising combustion mode, which increases fuel efficiency while also reducing NOx emissions to negligible levels. This Thesis concentrates on the implementation of this mode through experimental research, on an engine equipped with a fully variable valvetrain. Investigation of the operational window, emissions, fuel consumption, thermodynamic efficiency is carried out and ways to improve on these are discussed. The governing consideration, however, is the control method for this rather intricate combustion mode. As such, experimental data acquisition and analysis of ion current under the whole operating spectrum, from spark ignition to full autoignition is made. It is found that the expected gains in fuel consumption and emissions are realized. In addition, ion current proves to be a very powerful and cost effective tool for engine monitoring, diagnosis and control. The author concludes that Controlled Auto Ignition is a viable proposition for mass production engine designs and that ion current, although not absolutely vital for engine control, considerably increases engine control thus allowing for greater operating window under autoignition, without compromising reliability or cost.

The research described in this thesis is concerned with the use of information technology to reduce manufacturing system downtime. The work was sponsored by Ford Motor Company Ltd, and the pilot information system was evaluated and demonstrated in their Bridgend engine plant. The maintenance strategy followed in this factory is total productive maintenance (TPM). Downtime reduction can offer significant benefits to a manufacturer through increases in capacity at no extra cost. Many organisations have implemented computerised maintenance management systems (CMMS) to improve their maintenance business processes, but studies have shown that CMMS have often delivered few real benefits. Although CMMS can assist the management of maintenance, they are under threat in this respect from more generic software such as enterprise resource planning (ERP). Studies have shown that within manufacturing organisations most CMMS users are managers, but very seldom production operators. CMMS have been accused of being too technology-centred and may therefore be unsuitable where a human-centred philosophy such as TPM is used. A pilot hypermedia information system has been designed and evaluated with the collaboration of maintenance managers and the work force of the engine plant. A version of the system was developed to investigate the use of a digital manual for data capture. The results of usability trials suggest that hypermedia manuals will be accepted and used by factory personnel in TPM teams, and that such a manual can be easily updated to provide an asset history. A hypermedia manual has been shown to enable knowledge management since knowledge gained during the process life-cycle may be added to that provided in the manual by the process designers. The main benefit of a hypermedia maintenance manual is rapid access to information, which can reduce the duration of unplanned maintenance activities.

A combination of environmental awareness, consumer demands and pressure from legislators has led automotive manufacturers to seek for more environmentally friendly alternatives while still meeting the quality, performance and price demands of their customers. This has led to many complex powertrain designs being developed in order to produce vehicles with reduced carbon emissions. In particular, within the last decade most of the major automotive manufactures have either developed or announced plans to develop one or more hybrid vehicle models. This means that to be competitive and o er the best HEV solutions to customers, manufacturers have to assess a multitude of complex design choices in the most e cient way possible. Even though the automotive industry is adept at dealing with the many complexities of modern vehicle development; the magnitude of design choices, the cross coupling of multiple domains, the evolving technologies and the relative lack of experience with respect to conventional vehicle development compounds the complexities within the HEV design space. In order to meet the needs of e cient and exible HEV powertrain modelling within this design space, a parallel is drawn with the development of complex software systems. This parallel is both from a programmatic viewpoint where object-oriented techniques can be used for physical model development with new equation oriented modelling environments, and from a systems methodology perspective where the development approach encourages incremental development in order to minimize risk. This Thesis proposes a modelling method that makes use of these new tools to apply OOM principles to the design and development of HEV powertrain models. Furthermore, it is argued that together with an appropriate systems engineering approach within which the model development activities will occur, the proposed method can provide a more exible and manageable manner of exploring the HEV design space.The exibility of the modelling method is shown by means of two separate case studies, where a hierarchical library of extendable and replaceable models is developed in order to model the di erent powertrains. Ultimately the proposed method leads to an intuitive manner of developing a complex system model through abstraction and incremental development of the abstracted subsystems. Having said this, the correct management of such an e ort within the automotive industry is key for ensuring the reusability of models through enforced procedures for structuring, maintaining, controlling, documenting and protecting the model development. Further, in order to integrate the new methodology into the existing systems and practices it is imperative to develop an e cient means of sharing information between all stakeholders involved. In this respect it is proposed that together with an overall systems modelling activity for tracking stakeholder involvement and providing a central point for sharing data, CAE methods can be employed in order to automate the integration of data.

The flow around a generic car model both in isolation and in proximity to a near side wall has been investigated utilising experimental and computational methods. Phase one of this investigation tested a range of Ahmed generic road vehicle models with varying backlight angles in isolation, employing laser-Doppler anemometry, static pressure and aerodynamic force and moment measurements in the experimental section. Additionally, numerical simulations were conducted using a commercial Reynolds-averaged Navier Stokes (RANS) code with the RNG k-ε turbulence model. This phase served both to extend the previous knowledge of the flow around the Ahmed model, and analyse the effects of both the supporting strut and rolling road. Phase two then used similar methods to investigate the Ahmed model in proximity to a non-moving side wall. Results from phase two are compared with previous near-wall studies in order that an understanding of the effects of wall proximity can be presented, an area lacking in the existing literature. It is found that the flow on the isolated model must be understood before the effects of side wall proximity can be assessed. There is though, in general, a breakdown of any longitudinal vortices on the near-wall side of the model as model-to-wall distance reduces, with an increase in longitudinal vortex strength on the model side away from the wall. There also exists a large pressure drop on the near-wall model side, which increases in magnitude as model-to-wall distance reduces, before dissipating at separations where the boundary layer restricts the flow. Additionally, there is found to be a pressure drop on the top and bottom of the model with decreasing wall distance, with the relative magnitudes of these dependent on model geometry.

Physical tests were performed with a bicycle and a dummy in a controlled laboratory environment to reproduce cyclist accidents. The kinematics of 13 sled tests were used to identify the cyclist head impact location, understand the interaction between the cyclist and bicycle and to validate a mathematical model. The finite element software code LS-DYNA was used to simulate 70 cyclist and pedestrian accidents with motor vehicles with four different vehicle shapes which supplemented the physical testing. The study has shown that when cyclists and pedestrians were struck by any of the vehicles their whole body kinematics can be distinguished into two phases, initially a rotation followed by a sliding action. The Sports Utility Vehicle (SUV) vehicle produced more of a rotation action rather than sliding, whereas the other vehicles produced a combination of the two. The current pedestrian legislation does not cover all head impact locations for cyclists and therefore needs to be extended to encompass the windscreen and A-Pillar regions of the vehicles. The wrap around distance (WAD) for all the vehicles, apart from the SUV, should be extended to encompass a larger region. For the SUV the current WAD region is adequate in protecting cyclists and pedestrians and does not need to change. The predicted head impactor angle for cyclists is 40 degrees which is lower than the current legislative value of 65 degrees and the predicted pedestrian head impact angle is higher at a value of 80 degrees for the MPV, SM and LFC. For the SUV the proposed impactor angle increased to 100 degrees for cyclists and pedestrians. This research has demonstrated significant differences in terms of input variables and outcomes between cyclist and pedestrian accidents involving vehicles. It has used mathematical models to obtain injury data from a human mathematical model and physical testing to replicate real world cyclist accident scenarios. Recommendations have been proposed for future legislative testing techniques for cyclists, based on existing pedestrian legislation. These recommendations to alter legislation will improve vehicle design and make future vehicles more cyclist-friendly.

No description available.