Model-based diagnosability and sensor placement

It is commonly accepted that the requirements for maintenance and diagnosis should be considered at the earliest stages of design. For this reason, methods for analysing the diagnosability of a system and determining which instrumentation is needed to achieve the desired level of diagnosability, are highly valued. This paper presents a method for: • Assessing the degree of diagnosability of a system, i.e. given a set of sensors, which faults can be discriminated? • Characterising and determining the minimal additional sensors which guarantee a specified degree of diagnosability. This analysis of a given system can be performed at the design phase, allowing one to determine the which sensors are needed, or the trade off if not installing certain sensors. This method has been applied to several subsystems of a General Electric Frame 6 gas turbine owned by National Power CoGen, UK in the framework of the European Community Trial Application project, TIGER Sheba. This paper focuses on the gas fuel subsystem for illustrating the method.