Comparison of fuzzy modelling approaches for fault detection systems

Fuzzy modelling together with Parallel Distributed Compensation (PDC)-based system analysis and controller/observer design techniques have emerged among the methods for developing fault detection and isolation (FDI). This paper provides a comparison of fuzzy logic modelling approaches for FDI of complex systems. The work is motivated by the modelling issues of the Takagi-Sugeno (T-S) fuzzy inference operator-based approximation structure that is essential for the implementation of a PDC system. The paper focuses on an application study of fault diagnosis for an industrial actuator system, conducted within the framework of the European study DAMADICS.