A Hybrid Neuro-Fuzzy and De-Coupling Approach Applied to the DAMADICS Benchmark Problem

A novel multiple-model fault detection and isolation scheme for non-linear dynamic systems, that of a Neuro-Fuzzy Decoupling Fault Detection Scheme (NFDFDS) is presented, a hybrid scheme incorporating both neuro-fuzzy and model-based methods. The FDI scheme employs local optimal observers designed according to minimum state estimation variance. An application of FDI for an electro-pneumatic valve actuator in a sugar factory is presented. Key issues of finding a suitable structure for detecting and isolating nine realistic actuator faults are described.