TIGER: numeric and qualitative model based diagnosis

This work is part of the work of ESPRIT project TIGER. This is an almost-completed three-year project to develop a situation assessment system for real-time, hard-to-measure dynamic systems. We are interested in the monitoring and diagnosis of such dynamic systems. We use implicit behavioural models supporting model-based approaches. This work is driven by the needs of and applied to, two gas turbines of very different size and power. In this paper, we describe how we have developed a system to provide a comprehensive fault detection and diagnosis system. The system described is currently working continuously, online to a gas turbine in a major chemical plant