This project involves the application of confocal Raman microscopy to the study of ageing effects in polymeric materials. The work involves three elements: a study of the relationships between the optical characteristics of the sample material and the confocal sampling process; the use of Raman microscopy as a analytical tool for the study of defect structures (voids, trees, corona discharge etc) in dielectrics; an investigation of the chemical changes that occur during exposure to high electric fields.
Although much is known, phenomenologically, about the electrostatic charging behaviour of insulator surfaces, many fundamental aspects of the process remain unclear. In this study we are applying two complementary techniques to study the charging/discharging behaviour of model insulator surfaces - size effects are of particular interest.
It is well known that mechanical forces can impact upon dielectric breakdown behaviour. In this project we are examining a range of issues associated with thermo-mechanical processes in high voltage cable insulation, and related model systems.
This project is concerned with the association of small molecular penetrants with polymeric systems. A range of techniques are being used to study the migration of small molecules through systems that are relevant to high voltage cables systems. Much of our current interest centres on the use of dilute solution techniques to provide fundamental thermodynamic parameters and the effect of ageing on these.
The project is examining the use of expertise finders and capability mangers within manufcaturing industries. The main challenges of the work are:
Methodologies for collecting information regarding skills and competencies through mining documents
Understanding the social requirements of such a system
Development examplar tools to help users locate the required expertise
Vast quantities of plastic waste packaging materials such as bottles, food container etc are currently sent to land-fill. This 14-month project sponsored by the Onyx Environmental Trust investigates the use of electrostatic tehniques to identify the different plastic types which are then segregated from each other and recycled.
A pilot-scale conveyor line is under construction in the EPE Group laboratories which handles mixed waste and ejects the different plastics at different points on the line. 'Non plastics' such as cardboard and wood are also streamed out.
Control systems are described by equations (eg differential equations), but their properties of interest are most naturally expressed in terms of the system trajectories (the set of all solutions to the equations). This is formalized by the relatively new notion of the system behaviour, due to Willems. The manipulation of system equations on the other hand can be formalized using algebra, more precisely module theory for linear systems. The relationship between modules and behaviours is very rich and leads to deep results on system structure.
The aim of this project is to investigate this module-behaviour correspondence and apply it to deepen our understanding of control systems theory and address outstanding problems. We are particularly interested in the application area of multidimensional systems, i.e. systems described by partial differential equations or the discrete equivalent. In this area, we have to date had much success using these tools, eg in the characterization of controllability, the definition, characterization and decomposition of system poles, and the investigation of the relationship between feedback and trajectory control. Current areas of work include the extension to systems with variable coefficients, and the development of a general theory of model reduction and system identification for such systems.
A major aim of the project at Southampton is to develop a database for research work in automatic gait recognition. We describe elsewhere our own research in automatic gait recognition where many of our students apply their new moving-feature extraction and description techniques. The advantages of using gait as a biometric are that it is non-contact and sequence based.
The work proposed in this project seeks to investigate the issues involved in the application of agents in diverse platforms from mobile communication devices to consumer electronics, specifically in relation to enabling technologies for the construction of conceptually grounded, practical frameworks for agent systems development
The aim of the project is to implement Iterative Learning Control (ILC) on a multi-axis conveyor belt system, similar to those found in industry. Conveyor systems are widely used throughout industry to transport products from one industrial process to another. Typical applications can be found in the automotive and food production industries. Particular attention will be given to the food production industry where conveyor systems are generally based on two parallel lengths of chain with a conveying surface between them. PID Control is practically the only system used, and these systems could well be described as "rough and ready".
A section of chain conveyor will be contructed with a range of different dispensers along its length, totalling in the region of 8 axes of motion. ILC will be used to improve positioning accuracy of the conveyor in both synchronous and indexing modes of operation, with a focus on reducing the relative error between dispensor and product.