PhD Research Positions in Automotive and Aeronautics

Closing Date for Applications: Friday, 1st May 2009.
There are currently research students working in all areas of Aeronautical and Automotive research disciplines These being in: Acoustics and Vibration, Aerodynamics, Combustion and Energy Conversion, Dynamics and Control, Materials and Structures and Risk and Reliability.

There are a number of fully funded PhD positions available, the majority to start at the beginning of each academic year; however others are available throughout the year. For specific studentships currently available (and likely to be available) within the research disciplines then have a look at the Available Projects below

Available Projects (PhD)

Projects currently available, with funding, are identified in the table below with an F. Projects potentially available, subject to funding acquired, are identified with an A.

To find out more about each of the individual projects click on the project title. To discover more about the research carried out within the department please visit the individual research group web links.

If you are interested in taking part in the leading-edge research then please contact the staff member highlighted for further information. For funded projects, if you want to apply then please contact

Jackie Green
Research Administrator
AAE Department
Stewart Miller Building
Loughborough University
Loughborough, Leicestershire
LE11 3TU

PhD01: Industrially Supported PhD Studentship in the Application of Multi-Axis Reinforced Polymer Composites for Advanced Aerospace Structures

Industrial Partner: M. Wright and Sons

Applications are invited for an industrially supported doctoral studentship in the Department of Aeronautical and Automotive Engineering at Loughborough University , leading to the award of a PhD degree. The research will be concerned with the application of 3D woven textile composites for advanced aerospace structures, with an emphasis on the structural integrity and performance of aerospace structures manufactured using this type of advanced fibre architecture. The work will investigate how crimp in the fibres and resin pockets within the matrix affect the mechanical performance (including post-impact performance) , and will consist of both experimental and numerical modelling aspects.

The project is supported by M. Wright and Sons, who are specialist manufacturers of woven narrow fabrics and technical textiles. M. Wright and Sons will support the research through the use of their unique multi-dimensional weaving machine.

Studentships will be paid a tax-free stipend of £12,940 per annum, plus £2,500 per annum supporting company contribution, plus tuition fees at the UK/EU rate for 3 years. These awards are tax free, and the research student will be eligible for additional income from supervision of undergraduate tutorials and laboratories.

Eligibility:
Applicants should have a minimum of a 2:1 honours degree or equivalent in Aeronautical Engineering, Mechanical Engineering or related disciplines. Experience in composites fabrication and the use of standard FEA codes such as Nastran or Abaqus would be an advantage.

Additional Information:
Applicants should complete the standard research degree application form available from:

http://www.lboro.ac.uk/prospectus/pg/apply/index.html

Further information about the studentship can be obtained by contacting:- Jackie Green, Research Administrator, Department of Aeronautical and Automotive Engineering, Loughborough University, E-mail: J.A.Green@lboro. ac.uk

For informal enquiries contact Dr Paul Cunningham (E-mail: P.Cunningham@ lboro.ac. uk )
Closing Date for Applications: Friday, 1st May 2009.

EPSRC Industrial CASE Studentship
in collaboration with Rolls Royce

Numerical Simulation of Aeroengine
Installation Aerodynamics

Dept. of Aeronautical and Automotive Engineering, Loughborough University

http://www.lboro.ac.uk/departments/tt/

Applications are invited for a research studentship at Loughborough University, leading to the award of a PhD degree. We are able to offer an opportunity to work in a challenging field that will combine industrial application with substantial research challenge. The project will be conducted within the Rolls-Royce University Technology Centre in the Dept. of Aero & Auto. Engineering. The aim of the project is to develop, validate, and apply a range of CFD methods for the analysis of the exhaust ducts, nozzles and near-field plume flowfields relevant to novel aeroengine architectures currently being developed by Rolls-Royce. Initial work will focus on RANS CFD for prediction of the flow within aggressively curved and strongly 3D exhaust ducts. Subsequent work will investigate alternative closure methods, including Large Eddy Simulation (LES). Both RANS and LES CFD will be carried out primarily using the Rolls-Royce Hydra code. The work will involve
collaboration with a team of Loughborough researchers as well as Rolls-Royce engineers. Experimental data is being gathered within the team and will be available for validation purposes. There will also be placements at Rolls-Royce, totalling three months, with travel and subsistence costs paid.

Eligibility:
Candidates should hold, or expect to receive, a first or upper second degree in a relevant engineering subject. Experience of CFD for aerospace applications would be an advantage.

Funding includes a maintenance award provided at the standard EPSRC level (currently £12,940 per annum in 2008/09) plus a sponsoring company contribution of £2,500 per annum, plus payment of UK/EU fees for 3.5 years. These awards are tax free, and the research student will be eligible for additional income from supervision of undergraduate tutorials and laboratories.

The studentship is subject to the EPSRC rules for eligibility, see:

http://www.epsrc.ac.uk/PostgraduateTraining/StudentEligibility.html

It is intended that the studentship should commence no later than July 2009..

Additional Information:
Applicants should complete and submit the PhD application form:

http://www.lboro.ac.uk/prospectus/pg/apply/index.html

And return to: Jackie Green, Research Administrator, Department of Aeronautical and Automotive Engineering, Loughborough University , J.A.Green@lboro. ac.uk
For informal enquiries contact Dr Gary Page ( G.J.Page@lboro. ac.uk ) 01509 227205.

PhD03: Automation of a system reliability model

Research Group: Risk and Reliability

Project Description:

Over the years various mathematical models have been developed that assess the reliability of a system. For example, Fault Tree Analysis, Cause-Consequence Analysis, Markov methods and Monte Carlo Simulation. Such models relate the performance of a given system design to the performance of the components of which the system is comprised and can be used to determine the failure probability or failure frequency of the system in question. Ideally such models would be used at the system design stage in order to assess the effect upon reliability of different design proposals and maintenance strategies. In this way design modifications can be accommodated cost effectively. . However, this ideal is not usually achieved in that reliability assessment does not efficiently influence the design. One way of improving this situation is to automate the reliability analysis process. This would make the analysis less complex enabling it to be performed by the design
team. Automation also reduces the time of an analysis and can help prevent errors.

The aim of the proposed studentship is to develop a methodology, implemented as a piece of software, that will accept as input the description of an engineering system, in the form of a schematic diagram, and use the information to generate a model to assess the system reliability. It is intended that the Cause Consequence method is initially considered.

If interested please contact Sarah Dunnett for further information at:

http://www.lboro.ac.uk/departments/tt/staff/dunnett.html

Fault Diagnostic Methodologies for Complex Systems (3 Projects)

John Andrews

Background

Fault diagnostic systems monitor the performance of the primary system to determine when it is not functioning correctly and identify the potential component failures that can have caused the symptoms observed. The symptoms are provided by sensors installed on the primary system to track the status of system variables during its operation. Faults on a system can be of concern for two reasons. Firstly they may cause or contribute to the down-time of the system with the associated financial implications. Secondly if the system is safety critical a degraded level of functionality may cause or increase the likelihood of fatalities.

With modern complex systems the diagnosis of the causes of a failed or degraded system state can be a difficult and time consuming task. Tools to support this activity have obvious benefits. The fault diagnosis process to identify the failed components can take a substantial proportion of the time it takes to rectify failures.

Fault diagnostic methods are numerous and diverse. However despite considerable research on this topic and the wide variety of methodologies available it remains a relatively immature science with many problems still to be overcome to enable its confident, widespread application to complex systems in many industries.

The majority of techniques developed for fault diagnosis work successfully when the symptoms observed are assumed to result from a lone failed component. When multiple failures have occurred the problem becomes very much more complex with the combined effects of the failures at times being very different from the symptoms observed for each individual failure. Three combinatorial techniques conventionally used for system reliability prediction, fault tree analysis, digraphs, and Bayesian belief networks have been investigated to determine their suitability to form the foundation of a fault diagnostic process capable of overcoming these deficiencies.

All three methods successfully accounted for multiple faults. Bayesian belief networks are better able to handle the introduction of evidence from the sensors if their structure can be established. Difficulties in their construction was overcome by direct development from the fault trees. This provided a successful means of fault diagnosis when the system operates under steady state conditions and the conditions were assumed to exist upon the system start-up.

Research Projects

PhD04 - Project 1:

Real-time application of a fault diagnostic method

This research project would address the following issues:

1. Dynamics need to be taken into account as symptoms, from the same failure conditions will change over time. For example consider a tank with a flow outlet at its base which develops a leak. Initially the symptoms will be a decreasing level and a continued flow from the outlet. When the tank is empty the flow out will then cease.
2. The time duration between the occurrence of multiple faults . On the first fault occurrence all symptoms will be consistent with this fault alone. The occurrence of other failures may then change or mask these symptoms. The time duration between the occurrence of the multiple failure events will affect the symptoms observed. Successful diagnosis of the causes will require the consideration of the inter-arrival time of the faults.

PhD05 - Project 2:

Optimal Sensors for Fault Diagnosis

This research project would address the following issues:

It will be impractical to feature the number of sensors required to exactly isolate the precise component failures causing each system fault condition.. As such a list of potential failures will be deduced. The number of sensors installed in the system will be a trade-off between the cost of the sensors (plus their maintenance) and the value of the information they produce to the fault diagnostics process. The sensors used should optimise the value of the information provided with constraints placed on the resources available.

PhD06 - Project 3:

Large Scale System Fault Diagnostics

This research project would address the following issues:
The success of a general fault diagnostic methodology will depend on its ability to scale up to ever increasing system complexities. Larger systems will require more sensors to monitor its status providing the potential for information overload with the fault identification system unable to cope. As such the systems need to be modularised to generate a system structure where faults are progressively tracked down through sub-systems and then sub-sections giving, at each stage, a problem whose fidelity can be handled effectively. The modules need to be relatively independent and a method to identify these is required.