Foreseen Results

The proposed research will be carried out within the framework of industrial application. Therefore, the research will consider practical requirements in the real fault diagnosis situation. All fault diagnosis methods will be assessed using laboratory systems or data supplied by industry. The design software tools will be developed in association with all theoretical developments. Hence one of the main achievements of this project will be a number of design software tools. This will offer particularly important benefits to industry in that it will enable development engineering to evaluate and select diagnostic tools and techniques - facilitating "application roll-out" which is seen as the major problem preventing the commercial development of these technologies.
The aim of this project is to develop mechanisms for integrating new quantitative and qualitative methods within the framework of industrial applications and based upon an application study in a sugar factory.

The objectives are :

new research into quantitative & qualitative methodologies for robust fault diagnosis,
the development of a strategy for industrial application,
exchange of knowledge between partners. The project will operate with the advice of an industrial advisor, Dr. S. Daley.

This project is to develop advanced fault diagnosis techniques to increase overall process reliability. There is an increasing demand for industrial process to become more safe and reliable. This requirement extends beyond the normally accepted safety-critical systems of nuclear reactors, chemical plants and aircraft, to new systems such as autonomous vehicles and fast railways. The early detection of faults can help avoid system shut-down, breakdown and even catastrophes involving human fatality and material damage. The maintenance cost of industrial processes can therefore be reduced and product quality can be enhanced.

Quantitative fault diagnosis is a subject studied in control engineering, whilst qualitative fault diagnosis is normally studied within computer science. This proposed project, in bringing together an innovative marriage of these two different subjects to achieve highly reliable and efficient fault diagnosis, represents a significant advancement both in the theory and application of this important field on control engineering. The integrated approach represents the way forward into the next generation of fault diagnosis. One of the main achievements of this project will ne new theoretical developments of faults diagnosis methodologies which covers quantitative and qualitative approaches and their integration. The research on integration of quantitative and qualitative approaches will contribute not only to the development of the fault diagnosis field, but also to automatic control in general.